R. v. McDonnell, 1997 CanLII 389 (SCC), [1997] 1 SCR 948

Terry McDonnell Appellant

v.

Her Majesty The Queen Respondent

Indexed as: R. v. McDonnell

File No.: 24814.

1996: December 6; 1997: April 24.

Present: Lamer C.J. and La Forest, L’Heureux‑Dubé, Sopinka, Gonthier, Cory, McLachlin, Iacobucci and Major JJ.

on appeal from the court of appeal for alberta

Criminal law ‑‑ Sentencing ‑‑ Appeals ‑‑ Sexual assault ‑‑ Accused pleading guilty to two counts of sexual assault ‑‑ Trial judge sentencing accused to 12 months’ incarceration for first offence and to six months for second offence, to be served concurrently ‑‑ Court of Appeal finding assaults to be major sexual assaults and raising sentence for first offence to four years’ incarceration and sentence for second offence to one year, to be served consecutively ‑‑ Whether Court of Appeal erred in overturning sentences imposed by trial judge.

Criminal law ‑‑ Sentencing ‑‑ Appeals ‑‑ Applicable standard of review of sentencing decisions.

Criminal law ‑‑ Sentencing ‑‑ Sexual assault ‑‑ Starting‑point approach.

The accused pleaded guilty to two counts of sexual assault, contrary to s. 271 of the Criminal Code. The first offence occurred in 1986 when the accused was 29. The complainant, a 16‑year‑old girl, had been placed in the accused's home by Social Services. She was asleep on the living room couch when the accused came home intoxicated. He undid her pants. The complainant turned over on her stomach and tried to press herself into the couch, pretending to be sleeping. He took off her jeans, started to kiss her buttock area and partially penetrated her vagina with his penis. She had her legs pressed together and he was trying to get them apart. Finally, he said "You're too difficult" and rolled onto the floor where he fell asleep. The second offence occurred in 1993 when the accused was 36. The complainant was a 14‑year‑old babysitter for the accused's family. She had fallen asleep on her stomach on a hide‑a‑bed. She woke up in the middle of the night; her underwear was pulled down and the accused was on top of her, rubbing her back with one hand and her buttocks with the other. He also touched her pelvis and vaginal areas. She screamed and fled the house. The sentencing judge found that neither of the two assaults was a major sexual assault, as defined in past cases by the Court of Appeal of the province, with a starting‑point sentence of three years, and imposed a sentence of 12 months in custody for the first offence and six months for the second, to be served concurrently. The Court of Appeal allowed the Crown's appeal. The court found that the sexual assaults were major ones and sentenced the accused to four years in custody for the first offence and, considering the global effect of the sentences, to one year for the second, to be served consecutively.

Held (La Forest, L’Heureux‑Dubé, Gonthier and McLachlin JJ. dissenting): The appeal should be allowed.

Per Lamer C.J. and Sopinka, Cory, Iacobucci and Major JJ.: The Court of Appeal’s decision is inconsistent with the deference that is owed to sentencing judges by appellate courts. In the absence of an error of principle, failure to consider a relevant factor, or overemphasis of the appropriate factors, a sentence should only be overturned if it is demonstrably unfit. Here, the Court of Appeal failed to point to a relevant factor not considered by the sentencing judge that would give rise to appellate review of the sentence. With respect to the first offence, the sentencing judge clearly considered penetration and psychological trauma as factors in reaching a sentence and did not treat the other personal problems the first complainant had been having around the time of the assault as a mitigating factor. After taking into account all relevant mitigating and aggravating circumstances, the sentencing judge arrived at what she considered was an appropriate sentence. With respect to the second offence, the same conclusion applies: the sentencing judge did not fail to consider relevant factors.

The mischaracterization of an offence according to judicially created categories is not an error in principle, and is an inappropriate ground to set aside the lower court’s sentence. First, deference should be shown to a lower court’s sentencing decision. Second, there is no legal basis for the judicial creation of a category of offence within a statutory offence for the purposes of sentencing. It is not for judges to create criminal offences, but rather for the legislature to enact them. By creating a species of sexual assault known as a “major sexual assault”, and by basing sentencing decisions on such a categorization, the Alberta Court of Appeal has effectively created an offence, at least for the purposes of sentencing. Given Parliament’s intention to treat sexual assaults causing bodily harm ‑‑ which includes psychological harm ‑‑ under s. 272(c) of the Criminal Code, it is inappropriate to create a “major sexual assault”, which is based at least in part on the existence of harm to the victim pursuant to s. 271. If the prosecution is to be based on the harm to the victim, the accused should be charged under s. 272(c). It is not for the courts to establish a subset of offence within s. 271 that is based on harm. Moreover, to the extent that the Court of Appeal held that the Crown need not prove psychological harm in some instances, but rather such harm may be presumed, it was in error. If the Crown wishes to rely upon the existence of psychological harm, it should charge under s. 272(c), and prove the offence. If harm is accepted as an aggravating factor under s. 271 and psychological harm may be presumed, the burden of proving harm as an aggravating factor is improperly lifted from the Crown and a burden of disproving harm shifts to the accused. In this case, the Court of Appeal’s discussion of the presumption of harm was both erroneous and unnecessary; harm existed and was considered in setting the sentence.

The first sentence’s departure from the Court of Appeal’s view of the appropriate starting point of three years does not in itself imply that the sentence was demonstrably unfit. Nor was there any reason given by the Crown or the Court of Appeal to conclude that a six‑month sentence for the second offence was demonstrably unfit. An appellate court may set out starting‑point sentences as guides to lower courts and this starting point may well be a factor to consider in determining whether a sentence is demonstrably unfit. If there is a wide disparity between the starting point for the offence and the sentence imposed, then, assuming that the Court of Appeal has set a reasonable starting point, the starting point certainly suggests, but is not determinative of, unfitness. Unless there otherwise is a proper reason to interfere with the sentence, a sentence cannot be altered on appeal, notwithstanding deviation from a starting point.

The decision to order concurrent or consecutive sentences should be treated with the same deference owed by appellate courts to sentencing judges concerning the length of sentences ordered. The Court of Appeal failed to raise a legitimate reason to alter the order of concurrent sentences made by the sentencing judge; the court simply disagreed with the result of the sentencing judge’s exercise of discretion, which is an insufficient reason for interfering.

Per La Forest, L’Heureux‑Dubé, Gonthier and McLachlin JJ. (dissenting): A just sentence is one which reflects the seriousness of the crime and fits the individual circumstances of the accused. Appellate courts in a number of provinces have responded to the need to accommodate these dual goals and resolve the tension which may arise between them by introducing the concept of “starting‑point” sentences. The starting‑point approach to sentencing involves two steps. The first step consists of determining the appropriate range of sentence for an offence of this type in a typical case. At this stage, the inquiry is objective. The appropriate range is based on the general characteristics of the typical offence and on the assumption that the accused is a person of good character with no criminal record. Using that range as a starting point, the second step involves adjusting the sentence upward or downward on the basis of aggravating or mitigating factors. The judge must consider factors personal to the accused and the victim, and the actual consequences of the offence. In short, the sentence must be individualized to the particular crime and the particular accused before the court. This approach thus represents an attempt to marry in one sentencing principle the values of uniformity and individualization. With the starting‑point approach, in all cases the final sentence will be the least that is appropriate in the circumstances.

The starting‑point approach, properly understood and applied, does not violate s. 7 of the Canadian Charter of Rights and Freedoms. This approach does not relieve the Crown of its burden of establishing the appropriateness of severe punishments. Once the Crown has established that the accused is guilty of the offence charged, it must then establish that the offence, in all the circumstances, falls within a particular range. This fixes the starting point. In order to obtain a harsher sentence, the Crown must also establish aggravating circumstances. On the other hand, if mitigating factors are revealed, the sentence will be reduced from the starting point. Further, the starting‑point approach in sexual assault cases does not create a vague law which fails to delineate the area of risk as required by s. 7 of the Charter. The principle against vagueness applies to the question of what conduct is criminalized and not to sentencing ranges. The principle is satisfied if the law specifies the conduct prohibited and indicates a range of punishment. Even if the principle against vagueness did apply to sentencing ranges, the starting point here at issue would satisfy it. If anything, the starting‑point approach increases the knowledge of the risk faced by a prospective or convicted accused. The possibility that judges may differ on whether to consider a particular assault major does not attract the vagueness principle. Vagueness does not arise simply because the law is open to varying interpretations by the courts.

The Criminal Code creates a single offence of sexual assault which embraces a wide range of conduct. The subdivision by the Alberta Court of Appeal of this vast offence into minor or major sexual assault for the purpose of sentencing, with a starting point of three years for major sexual assaults, does not amount to judicial legislation of a new offence. The starting‑point approach is merely a variation on the traditional concept of ranges of sentence for particular types of criminal acts. To recognize a certain type of act as being serious or major, and hence, in the typical case, attracting a sentence in a particular range, is not to create a new crime.

The starting‑point approach, properly conceived, does not involve a presumption of harm in sexual assault cases. When determining the starting point, the judge must consider whether the violation of the victim’s integrity was “such that a reasonable person would know beforehand that the victim likely would suffer lasting emotional or psychological injury, whether or not physical injury occurs”. The inquiry at that stage is concerned not with the actual facts of the case but with the sort of harm this type of assault would be likely to cause. Actual harm becomes an issue only at the stage of the individualization of the sentence where the judge must determine whether in fact the crime caused lasting emotional or psychological harm to the victim. At both stages, there is no presumption of harm. Because the starting‑point approach does not presume actual harm to the victim, it does not violate the rule that the Crown bears the burden of proving exacerbating circumstances, nor does it conflict with the logic underlying s. 272 of the Criminal Code, which requires proof of actual harm in the particular case.

Courts of appeal may intervene where the sentence discloses error of principle or is demonstrably unfit. The starting‑point approach does not provide a new judge‑made legal principle, enabling courts of appeal to interfere with the proper exercise of the sentencing judge’s decision. The starting point is not a principle of law, but rather a tool to determine the proper range of sentence for a certain type of offence. Failure to allude to the appropriate starting point or range is not an error of principle. When a court of appeal interferes on the ground that the judge ignored the correct starting point, it is simply saying that the sentence is demonstrably unfit because it falls outside the acceptable range of sentence for that sort of offence.

Here, the sentence imposed at trial was “demonstrably unfit”. A review of many cases from Alberta reveals that a sentence of less than two years in circumstances similar to the case at bar constitutes a marked departure from the norm. The unfit nature of the sentence resulted from the sentencing judge’s failure to place appropriate weight on the appropriate factors, thus depreciating the seriousness of the offence. Most significantly, she placed great emphasis on the physical acts but did not ask herself whether this was the sort of offence which would likely result in lasting psychological or emotional harm ‑‑ the hallmark of a major sexual assault. The Court of Appeal correctly identified the first offence as a major sexual assault. While it involved no violence or threats, the result of such an assault on a typical victim would likely have been lasting emotional and psychological harm. The case exhibited few mitigating factors but a number of aggravating ones, including abuse of trust and lasting emotional and psychological harm. By adding a year to the starting‑point sentence, the Court of Appeal imposed a sentence which reflected the seriousness of the offence as well as the particular circumstances of the accused. The second offence, viewed in isolation, also qualifies as a major sexual assault. The Court of Appeal correctly concluded that concurrent sentences were not appropriate in this case, which involved two separate offences on two different victims, separated by a period of seven years. Nevertheless, the global effect of the sentences must be considered and the Court of Appeal correctly imposed a sentence of one year for the second offence, to be served consecutively.

Cases Cited

By Sopinka J.

Applied: R. v. Shropshire, 1995 CanLII 47 (SCC), [1995] 4 S.C.R. 227; R. v. M. (C.A.), 1996 CanLII 230 (SCC), [1996] 1 S.C.R. 500; distinguished: R. v. S.G.O.R. (1991), 113 A.R. 36; R. v. S. (W.B.); R. v. P. (M.) (1992), 1992 CanLII 2761 (AB CA), 73 C.C.C. (3d) 530; R. v. Spence (1992), 1992 ABCA 352, 78 C.C.C. (3d) 451; R. v. Nicholson (1993), 1993 ABCA 248, 145 A.R. 262; R. v. Lapatak (1995), 1995 ABCA 322, 169 A.R. 385; referred to: R. v. Sandercock (1985), 1985 ABCA 218, 22 C.C.C. (3d) 79; R. v. A.B.C. (1991), 1991 ABCA 337, 120 A.R. 106; Frey v. Fedoruk, 1950 CanLII 24 (SCC), [1950] S.C.R. 517; R. v. McCraw, 1991 CanLII 29 (SCC), [1991] 3 S.C.R. 72; R. v. Gardiner, 1982 CanLII 30 (SCC), [1982] 2 S.C.R. 368; R. v. R.P.T. (1983), 1983 ABCA 175, 7 C.C.C. (3d) 109.

By McLachlin J. (dissenting)

R. v. Hessam (1983), 1983 ABCA 69, 43 A.R. 247; R. v. Sandercock (1985), 1985 ABCA 218, 22 C.C.C. (3d) 79; R. v. Natanson (1927), 1927 CanLII 518 (SK CA), 49 C.C.C. 89; R. v. Connor and Hall (1957), 1957 CanLII 509 (ON CA), 118 C.C.C. 237; R. v. Baldhead, 1966 CanLII 335 (SK CA), [1966] 4 C.C.C. 183; R. v. Jourdain and Kudyba (1958), 1958 CanLII 445 (MB CA), 121 C.C.C. 82; R. v. Zong (1986), 1986 CanLII 6903 (NS CA), 173 A.P.R. 432; R. v. Muswagon (1993), 1993 CanLII 14904 (MB CA), 88 Man. R. (2d) 319; R. v. Post (1996), 72 B.C.A.C. 312; R. v. Jackson (1993), 1993 CanLII 4414 (SK CA), 87 C.C.C. (3d) 56; R. v. Glassford (1988), 1988 CanLII 7094 (ON CA), 27 O.A.C. 194; R. v. Cunningham (1996), 1996 CanLII 1311 (ON CA), 104 C.C.C. (3d) 542; R. v. Edwards; R. v. Brandy, The Times, July 1, 1996; R. v. Jabaltjari (1989), 46 A. Crim. R. 47; Reference re ss. 193 and 195.1(1)(c) of the Criminal Code, 1990 CanLII 105 (SCC), [1990] 1 S.C.R. 1123; R. v. Shropshire, 1995 CanLII 47 (SCC), [1995] 4 S.C.R. 227; R. v. M. (C.A.), 1996 CanLII 230 (SCC), [1996] 1 S.C.R. 500; R. v. Muise (1994), 1994 NSCA 198, 94 C.C.C. (3d) 119; R. v. S.G.O.R. (1991), 113 A.R. 36; R. v. Sand (1991), 1991 CanLII 13091 (AB CA), 120 A.R. 397; R. v. S. (W.B.); R. v. P. (M.) (1992), 1992 CanLII 2761 (AB CA), 73 C.C.C. (3d) 530; R. v. R.S.B. (1992), 1992 ABCA 258, 135 A.R. 23; R. v. Spence; R. v. F. (D.L.) (1992), 1992 ABCA 352, 78 C.C.C. (3d) 451; R. v. R.E.C., [1993] A.J. No. 303 (QL); R. v. D.M.B. (1993), 1993 CanLII 16356 (AB CA), 141 A.R. 307; R. v. Nicholson (1993), 1993 ABCA 248, 145 A.R. 262; R. v. Dionne, [1993] A.J. No. 939 (QL); R. v. D.K. (1994), 1994 ABCA 219, 155 A.R. 269; R. v. Watson (1994), 1994 ABCA 234, 157 A.R. 80; R. v. Wagar (1995), 1995 ABCA 413, 174 A.R. 317; R. v. Lapatak (1995), 1995 ABCA 322, 169 A.R. 385; R. v. Lakotos (1996), 1996 ABCA 262, 187 A.R. 45; R. v. A.B.C. (1991), 1991 ABCA 337, 120 A.R. 106.

Statutes and Regulations Cited

Act to amend the Criminal Code (sentencing) and other Acts in consequence thereof, S.C. 1995, c. 22, s. 6.

Canadian Charter of Rights and Freedoms, s. 7.

Criminal Code, R.S.C., 1985, c. C‑46, ss. 271 [am. c. 19 (3rd Supp.), s. 10], 272, 724(3)(e) [repl. 1995, c. 22, s. 6].

Authors Cited

Canada. Sentencing Commission. Sentencing Reform: A Canadian Approach. Ottawa: The Commission, 1987.

Cross, Rupert, Sir. The English Sentencing System, 2nd ed. London: Butterworths, 1975.

Linden, Allen M. “A Fresh Approach to Sentencing in Canada”. In Hélène Dumont, ed., Sentencing. Cowansville, Que.: Canadian Institute for the Administration of Justice, 1987, 53.

Marshall, Patricia. “Sexual Assault, The Charter and Sentencing Reform” (1988), 63 C.R. (3d) 216.

Ruby, Clayton C. Sentencing, 4th ed. Toronto: Butterworths, 1994.

Stuart, Don. Charter Justice in Canadian Criminal Law. Scarborough, Ont.: Carswell, 1991.

Young, Alan. The Role of an Appellate Court in Developing Sentencing Guidelines. Ottawa: Minister of Supply and Services Canada, 1988.

APPEAL from a judgment of the Alberta Court of Appeal (1995), 1995 ABCA 200, 169 A.R. 170, 97 W.A.C. 170, allowing the Crown’s appeal against sentence. Appeal allowed, La Forest, L’Heureux‑Dubé, Gonthier and McLachlin JJ. dissenting.

Marvin R. Bloos, for the appellant.

Paul L. Moreau, for the respondent.

The judgment of Lamer C.J. and Sopinka, Cory, Iacobucci and Major JJ. was delivered by

1Sopinka J. -- This appeal concerns the decision of the Court of Appeal to overturn the sentences imposed by the sentencing judge for two counts of sexual assault simpliciter to which the appellant had pleaded guilty. The Court of Appeal, concluding that the first assault complained of was a “major sexual assault” as that court had defined it in past cases, set aside the global sentence of one year, and instead imposed a global sentence of five years.

2In my view, the decision of the Court of Appeal was inconsistent with the deference that is owed to sentencing judges by appellate courts. A disagreement over the categorization of the assault as the appellate court has defined the category is an inappropriate ground to set aside the lower court’s sentence. In the present case, the sentencing judge did not commit an error in principle, did not ignore relevant factors and did not impose a demonstrably unfit sentence. Accordingly, I would allow the appeal and reinstate the sentence imposed initially by the sentencing judge.

I. Facts

3Following a preliminary inquiry, the appellant pleaded guilty to two counts of sexual assault, contrary to s. 271 of the Criminal Code, R.S.C., 1985, c. C-46. The first offence occurred in 1986 when the appellant was 29 years of age. The complainant was then a 16-year-old ward of the Alberta government and had been placed in the appellant's home by Social Services. Approximately two weeks after her arrival in the appellant's home, the complainant was asleep on the living room couch, lying on her back. The appellant came home intoxicated. He undid the complainant’s pants, at which point the complainant turned over on her stomach and tried to press herself into the couch. She testified that she did so in the hope that perhaps he would go away if he thought she were sleeping. He did not. Instead, he took off her jeans and started to kiss her buttock area. She was trying to pretend that she was sleeping. Then he penetrated her vagina with his penis. She described the degree of penetration as being "a little bit" since she had her legs pressed together and he was trying to get them apart. Finally, he said "You're too difficult" and rolled onto the floor. He tried to put the complainant's jeans back on. The complainant waited until she was sure that the appellant was asleep on the floor and then she went downstairs to her room. At no time did the complainant consent to the appellant's actions.

4The second offence occurred in 1993 when the appellant was 36 years of age. The complainant was a 14-year-old babysitter for the appellant's family. She had fallen asleep on her stomach on a hide-a-bed while babysitting. She testified that when she had gone to bed, she had been wearing a T-shirt, underwear and had a sleeping bag wrapped around her. When she awoke at 3:30 a.m., her underwear was pulled down and the sleeping bag was around her feet. The appellant was on top of her, rubbing her back under her T-shirt with one hand and rubbing her buttocks with the other. He then reached under her stomach touching her pelvis and vaginal areas. The appellant was trying to turn her over onto her stomach. He did not touch her breasts. She screamed and fled from the residence.

5The appellant pleaded guilty to the two offences. The sentencing judge found that neither of the two sexual assaults was a major one. She imposed a sentence of 12 months in custody followed by two years’ probation. The Court of Appeal allowed the Crown's appeal, finding that the sexual assaults were major ones and sentenced the appellant to five years’ imprisonment: (1995), 1995 ABCA 200, 169 A.R. 170, 97 W.A.C. 170. The appellant appeals the Court of Appeal’s decision.

II. Relevant Statutory Provisions

[6] Criminal Code, R.S.C., 1985, c. C-46

271. (1) Every one who commits a sexual assault is guilty of

(a) an indictable offence and is liable to imprisonment for a term not exceeding ten years; or

(b) an offence punishable on summary conviction.

272. Every one who, in committing a sexual assault,

(c) causes bodily harm to the complainant ...

is guilty of an indictable offence and liable to imprisonment for a term not exceeding fourteen years.

III. Prior Judgments

A. Provincial Court of Alberta

7Given the importance of the specific findings of the sentencing judge to the issues in the present appeal, I set out here the reasons of Burch Prov. Ct. J. pertaining to sentence in their entirety:

I have reviewed the facts as provided to me by the Crown today and confirmed by Mr. Tatarchuk and as heard by me at the preliminary inquiry with respect to the charges laid by [the first complainant]. I find that neither the assault on [the first complainant] nor on [the second complainant] were major sexual assaults as described in Regina v. Sandercock, or even as later refined since that case.

The assault in 1986 against [the first complainant] occurred on a 16-year-old girl who was placed in Mr. McDonnell's home by Social Services to help her resolve problems in her own home. She liked and trusted the McDonnells and had nowhere else to go.

The assault, while reprehensible, was an isolated one, and it was a situation of far more than fondling as the accused attempted penetration. However, in that case, there was no involvement of violence nor of threats. It did not involve oral sex nor anal intercourse, and there was only partial penetration. The event was spontaneous due to Mr. McDonnell's drunkenness on the night in question. It was a traumatic experience for the victim, but she was already 16 years old and was having other problems which may have contributed to her subsequent state of mind.

She cared for the defendant and she cared for his wife, and thus she didn't report the incident, unfortunately so, or Mr. McDonnell might have subsequently quit drinking, thus avoiding the later repetition of the offence in circumstances of a similar nature.

In sentencing Mr. McDonnell regarding this first charge, I am cognizant of the decision in R. v. R.P.T. [(1983), 1983 ABCA 175, 7 C.C.C. (3d) 109 (Alta. C.A.)], but this is not a case where simple rehabilitation will suffice. Mr. McDonnell needs to commit to never drinking again, thus satisfying the concern that he may never re-offend if he does so only when drinking, but there must be some element of denunciation and general deterrence in this matter as well.

Thus, as stated in Regina v. R.P.T. at page 114, the only solution, however imperfect, is to graft a rehabilitative sentence to a denunciatory sentence.

In sentencing Terry McDonnell, I take into account his strong family support, the strong community support, his remorse and his desire to quit drinking, but also the trauma suffered by the victim at a time when she was already troubled, and the fact that F.A.C.S. doesn't see counselling as being of any use to prevent re-offending, for Mr. McDonnell might simply re-offend if drunk again.

Lastly, I recognize that the time elapsed since the offence has relevance in relation to the relative effect of this on both the accused and the victim.

I believe that a fit sentence is 12 months in custody followed by two years probation.

Regarding the charges of assault against [the second complainant], the victim in this case has been traumatized, but the acts of the accused were very much in the ‘less grave’ category. Mr. McDonnell is a man of otherwise good character and is a strong member of his community. He has always maintained employment and supported his family. An additional lengthy consecutive custodial sentence to the custodial sentence imposed on the first charge would only seek to destroy the accused and his family and is not necessary to deter others from committing such an offence.

I will thus sentence him to six months in gaol concurrent to the first sentence, plus probation for the same period of time.

B. Court of Appeal for Alberta (The Court)

8The Court of Appeal held that the sentencing judge erred in concluding that the first assault was not a major sexual assault. R. v. Sandercock (1985), 1985 ABCA 218, 22 C.C.C. (3d) 79 (Alta. C.A.), held (at p. 84) that the key to a major sexual assault is the “evident blameworthiness of the offender”, as reflected in the extent to which the offender’s actions demonstrated “contemptuous disregard for the feelings and personal integrity of the victim”. In the present case, the court held that the appellant breached the trust reposed in him as a foster parent and did so in a manner which involved a serious violation of the complainant’s sexual and personal integrity and demonstrated a complete lack of concern for the feelings of the young girl. There was partial penile penetration of the vagina.

9With respect to the defence argument that the Crown had failed to prove the psychological harm aspect of a major sexual assault, the court stated (at p. 173):

The first point we wish to make is that we cannot envision a situation where nonconsensual intercourse vaginal, anal or oral would not fall into the major sexual assault category. These acts, by their very nature, constitute the ultimate violation of a woman's sexual integrity. And where one is dealing with assaults falling into the major category because of their very nature and here we include rape, attempted rape, fellatio, cunnilingus, and buggery the violence, force and intensity of the act are a given: R. v. McCraw.... In addition, in each case, there also exists a very real likelihood of psychological harm. Therefore, what must be understood is that it is not necessary that the Crown prove the existence of this kind of harm as a condition precedent to the courts classifying a sexual assault as a major one. Psychological harm is presumed in the absence of evidence to the contrary. That is what this court said in Sandercock.

The court elaborated (at pp. 174-75):

To put the matter another way, the offender is being sentenced on the basis of a major sexual assault, not because any specific psychological consequences have flowed from the attack but rather because of the nature of the attack and the fact that it poses the very real likelihood of long-term emotional or psychological harm. The fact that no such harm may materialize, a fact one could not possibly know until the victim's life had been lived in its entirety, is not a mitigating factor. However, that said, this does not mean that the consequences of the sexual assault are irrelevant. The degree of seriousness of the actions may be measured against the likely long-term consequences of the prohibited act. In other words, where the psychological harm has been severe, that may well be an aggravating factor. Of course, where harm beyond that which would be normally presumed is claimed in a case, the Crown must lead evidence to substantiate it.

For these reasons, therefore, to bring into question what is presumed to be the psychological harm flowing from a major sexual assault, it is not enough that the defence simply denies that it exists. Bringing the presumed harm into question in this context means pointing to cogent evidence to the contrary. No such evidence existed in this case. [Emphasis in original.]

10The court rejected the argument that the first victim did not suffer trauma because she already had other problems as both incorrect at law and perverse; her problems made her even more vulnerable. In any event, the victim testified at trial that the assault caused her psychological harm and the victim impact statement, which was before the sentencing judge by consent, also indicated such harm.

11The court also held that the second sexual assault was serious, not minor. Given the nature of the attack, psychological harm is rightly presumed. The sentencing judge found that the victim was traumatized, although the judge acknowledged that the assault was in the “less grave” category.

12The court held that the gap in time between the two incidents was not a mitigating factor, nor was this a case where concurrent sentences were appropriate.

13The court held that it was satisfied that a fit and proper sentence was four years for the first sexual assault and, considering the global effect of the sentences, one year consecutive for the second assault. The court recommended consideration for early parole.

IV. Analysis

A. Deference to Sentencing Decisions

14This appeal concerns the sentencing of the appellant who had pleaded guilty to two counts of sexual assault simpliciter. The sentencing judge held that an appropriate sentence for the offence that occurred in 1986 was 12 months in custody with two years’ probation, while the second offence brought a sentence of six months in jail, concurrent with the one-year sentence, plus probation for the same period. The Court of Appeal raised the custodial sentences for each offence to four years and one year respectively, the second sentence to be served consecutively. To assess the validity of the Court of Appeal’s judgment, it is first necessary to establish the standard of appellate review of sentencing decisions.

15Two recent cases, R. v. Shropshire, 1995 CanLII 47 (SCC), [1995] 4 S.C.R. 227, and R. v. M. (C.A.), 1996 CanLII 230 (SCC), [1996] 1 S.C.R. 500, set out the applicable standard of review of sentencing decisions. Iacobucci J., writing for the Court, stated in Shropshire at paras. 45-50:

Section 687(1) reads as follows:

687. (1) Where an appeal is taken against sentence, the court of appeal shall, unless the sentence is one fixed by law, consider the fitness of the sentence appealed against, and may on such evidence, if any, as it thinks fit to require or to receive,

(a) vary the sentence within the limits prescribed by law for the offence of which the accused was convicted; or

(b) dismiss the appeal.

The question, then, is whether a consideration of the "fitness" of a sentence incorporates the very interventionist appellate review propounded by Lambert J.A. With respect, I find that it does not. An appellate court should not be given free reign to modify a sentencing order simply because it feels that a different order ought to have been made. The formulation of a sentencing order is a profoundly subjective process; the trial judge has the advantage of having seen and heard all of the witnesses whereas the appellate court can only base itself upon a written record. A variation in the sentence should only be made if the court of appeal is convinced it is not fit. That is to say, that it has found the sentence to be clearly unreasonable.

I would adopt the approach taken by the Nova Scotia Court of Appeal in the cases of R. v. Pepin (1990), 1990 CanLII 2481 (NS CA), 98 N.S.R. (2d) 238, and R. v. Muise (1994), 1994 NSCA 198, 94 C.C.C. (3d) 119. In Pepin, at p. 251, it was held that:

... in considering whether a sentence should be altered, the test is not whether we would have imposed a different sentence; we must determine if the sentencing judge applied wrong principles or (if) the sentence is clearly or manifestly excessive.

Further, in Muise it was held at pp. 123‑24 that:

In considering the fitness of a sentence imposed by a trial judge, this court has consistently held that it will not interfere unless the sentence imposed is clearly excessive or inadequate....

The law on sentence appeals is not complex. If a sentence imposed is not clearly excessive or inadequate it is a fit sentence assuming the trial judge applied the correct principles and considered all relevant facts.... My view is premised on the reality that sentencing is not an exact science; it is anything but. It is the exercise of judgment taking into consideration relevant legal principles, the circumstances of the offence and the offender. The most that can be expected of a sentencing judge is to arrive at a sentence that is within an acceptable range. In my opinion, that is the true basis upon which Courts of Appeal review sentences when the only issue is whether the sentence is inadequate or excessive.

Unreasonableness in the sentencing process involves the sentencing order falling outside the "acceptable range" of orders; this clearly does not arise in the present appeal. An error of law involves a situation such as that found in R. v. Chaisson, 1995 CanLII 58 (SCC), [1995] 2 S.C.R. 1118, in which a sentencing judge, while calculating the total time period of incarceration for the purposes of a "half‑time" parole ineligibility order under s. 741.2 of the Code, erroneously included two offences in the calculations notwithstanding the fact that these specific offences were not listed in the schedule of offences to which the s. 741.2 orders apply. [Emphasis in original.]

16The deferential approach set out in Shropshire was confirmed and refined in M. (C.A.). In that case, Lamer C.J., on behalf of the Court, stated at paras. 90-92:

Put simply, absent an error in principle, failure to consider a relevant factor, or an overemphasis of the appropriate factors, a court of appeal should only intervene to vary a sentence imposed at trial if the sentence is demonstrably unfit. Parliament explicitly vested sentencing judges with a discretion to determine the appropriate degree and kind of punishment under the Criminal Code. As s. 717(1) reads:

717. (1) Where an enactment prescribes different degrees or kinds of punishment in respect of an offence, the punishment to be imposed is, subject to the limitations prescribed in the enactment, in the discretion of the court that convicts the person who commits the offence.

This deferential standard of review has profound functional justifications. As Iacobucci J. explained in Shropshire, at para. 46, where the sentencing judge has had the benefit of presiding over the trial of the offender, he or she will have had the comparative advantage of having seen and heard the witnesses to the crime. But in the absence of a full trial, where the offender has pleaded guilty to an offence and the sentencing judge has only enjoyed the benefit of oral and written sentencing submissions (as was the case in both Shropshire and this instance), the argument in favour of deference remains compelling. A sentencing judge still enjoys a position of advantage over an appellate judge in being able to directly assess the sentencing submissions of both the Crown and the offender. A sentencing judge also possesses the unique qualifications of experience and judgment from having served on the front lines of our criminal justice system. Perhaps most importantly, the sentencing judge will normally preside near or within the community which has suffered the consequences of the offender's crime. As such, the sentencing judge will have a strong sense of the particular blend of sentencing goals that will be "just and appropriate" for the protection of that community. The determination of a just and appropriate sentence is a delicate art which attempts to balance carefully the societal goals of sentencing against the moral blameworthiness of the offender and the circumstances of the offence, while at all times taking into account the needs and current conditions of and in the community. The discretion of a sentencing judge should thus not be interfered with lightly.

Appellate courts, of course, serve an important function in reviewing and minimizing the disparity of sentences imposed by sentencing judges for similar offenders and similar offences committed throughout Canada.... But in exercising this role, courts of appeal must still exercise a margin of deference before intervening in the specialized discretion that Parliament has explicitly vested in sentencing judges. It has been repeatedly stressed that there is no such thing as a uniform sentence for a particular crime.... Sentencing is an inherently individualized process, and the search for a single appropriate sentence for a similar offender and a similar crime will frequently be a fruitless exercise of academic abstraction. As well, sentences for a particular offence should be expected to vary to some degree across various communities and regions in this country, as the "just and appropriate" mix of accepted sentencing goals will depend on the needs and current conditions of and in the particular community where the crime occurred. For these reasons, consistent with the general standard of review we articulated in Shropshire, I believe that a court of appeal should only intervene to minimize the disparity of sentences where the sentence imposed by the trial judge is in substantial and marked departure from the sentences customarily imposed for similar offenders committing similar crimes. [Emphasis in original.]

17I have included extensive references to these cases because in my view they are highly significant to the case at bar. M. (C.A.) set out that, in the absence of an error of principle, failure to consider a relevant factor, or overemphasis of the appropriate factors, a sentence should only be overturned if the sentence is demonstrably unfit. The respondent submitted that the sentencing judge in the present case failed to consider relevant factors and that the sentence was demonstrably unfit. Moreover, both the respondent and the Court of Appeal appear to have treated the failure of the sentencing judge to characterize the offence as a major sexual assault as an error in principle. I will discuss these contentions in turn.

B. Relevant Factors and Demonstrable Unfitness

18Sandercock, supra, established in Alberta the notion of a “major sexual assault”, which carried with it a presumptive sentence (“starting point”) of three years. Sandercock stated at p. 84 that the key to a major sexual assault is the “evident blameworthiness of the offender” as reflected in the extent to which the offender’s actions demonstrated a “contemptuous disregard for the feelings and personal integrity of the victim”. The Court of Appeal held in the present case that the sentencing judge erred in failing to find that the first offence amounted to a major sexual assault.

19In concluding that the sentencing judge had mischaracterized the nature of the assault which was the subject of the first offence, the Court of Appeal relied on several factors. First, the court stated (at p. 173):

One salient fact cannot be overlooked. This was not, despite the defence suggestion, a case of “fondling”. It was a case of penile penetration of the vagina. The fact that McDonnell only succeeded in partially penetrating the complainant’s vagina with his penis because of the complainant’s efforts to resist him does not make this any less a major sexual assault. Partial penetration will suffice. Accordingly, this assault falls squarely within what is described in Sandercock as one of the archetypical cases of major sexual assault.

Second, the court did not accept the submission of the defence, which the court stated (at p. 173) the “trial judge appears to have implicitly accepted”, that the first offence could not constitute major sexual assault because of the absence of psychological harm. The court held that non-consensual intercourse leads to a very high likelihood of trauma, which likelihood is one of the indicia of a major sexual assault according to Sandercock. In any event, the court concluded based on viva voce evidence and the victim impact statement that the complainant in the first case did suffer psychological harm. On the basis of these factors, the court concluded that the first offence was a major sexual assault and that the sentence ordered by the sentencing judge was insufficient.

20In my view, the Court of Appeal fails to point to a relevant factor not considered by the sentencing judge that would give rise to appellate review of the sentence. The first factor emphasized by the Court of Appeal, partial penetration, was explicitly cited by the sentencing judge. She stated:

The assault, while reprehensible, was an isolated one, and it was a situation of far more than fondling as the accused attempted penetration. However, in that case, there was no involvement of violence nor of threats. It did not involve oral sex nor anal intercourse, and there was only partial penetration. [Emphasis added.]

Clearly, the sentencing judge did consider penetration as a factor in reaching a sentence. Thus, consideration of this factor fails to give grounds to alter the sentence.

21The second factor alluded to by the Court of Appeal, psychological trauma, was also considered by the sentencing judge. The judge stated:

It was a traumatic experience for the victim, but she was already 16 years old and was having other problems which may have contributed to her subsequent state of mind.

The judge later stated:

In sentencing Terry McDonnell, I take into account his strong family support, the strong community support, his remorse and his desire to quit drinking, but also the trauma suffered by the victim at a time when she was already troubled, and the fact that F.A.C.S. doesn’t see counselling as being of any use to prevent re-offending, for Mr. McDonnell might simply re-offend if drunk again.

It is clear, in my view, that the sentencing judge did not fail to consider the trauma to the complainant in the first assault.

22Finally, the Court of Appeal stated (at p. 175) that it was “perverse” for the sentencing judge to treat the other personal problems the first complainant had been having around the time of the assault as a mitigating factor. I disagree with this characterization of the sentencing judge’s views in the matter. It appears to me that in the first statement above the sentencing judge noted the problems the complainant had been having as a partial explanation of her personal problems after the assault; that is, not all her problems after the assault were attributable to the assault. The second statement by the sentencing judge indicates that while there were mitigating factors, the complainant’s personal problems actually made the assault more serious. The Court of Appeal did not interpret the judge correctly, in my view, in reaching its conclusion that the sentencing judge had misused the evidence of the problems that the first complainant had been having.

23The respondent’s submission that the judge failed to consider relevant factors in my view cannot succeed with respect to the first offence. The respondent also submits that the sentence imposed was demonstrably unfit. The sentence originally imposed for the first offence was one year, whereas the Court of Appeal imposed a four-year sentence. The “starting point” as set out in Sandercock for a major sexual assault was three years. These differences in themselves provide me no basis to conclude that the judge’s sentence originally passed was demonstrably unfit.

24Sandercock does not purport to create a rigid tariff. At pp. 82-83 the Alberta Court of Appeal stated:

...the [sentencing] guidance offered should not be too rigid. A fixed guideline, or tariff (or, indeed, even an “approved range”), fails to take into account the immense variety of circumstances which can be found in different cases involving a conviction for the same offence. Even putting aside the offender’s circumstances, those who advocate some form of fixed sentences fail to appreciate that the definitions of the crimes in the Criminal Code contain only certain key elements required for guilt.... The manifest object of the Criminal Code is that the sentencing process will adjust for the other important factors, whether aggravating or mitigating. This is why the sentencing judge is given a wide scope of terms of possible sentences. [Emphasis added.]

Indeed, for reasons which follow, I conclude that it would be inappropriate to do so. Faithful to this instruction, the sentencing judge took into account all relevant mitigating and aggravating circumstances and arrived at what she considered was an appropriate sentence. Accordingly, the sentence’s departure from the Court of Appeal’s view of the appropriate starting point does not in itself imply that the sentence was demonstrably unfit.

25Moreover, I note that in a case not dissimilar to the present case, the Alberta Court of Appeal imposed a custodial sentence of one year. In R. v. A.B.C. (1991), 1991 ABCA 337, 120 A.R. 106, the accused had sexually assaulted his sedated 16-year-old daughter, fondling her breasts and vagina and possibly penetrating her vagina with his penis. The Court of Appeal vacated the two-year suspended sentence and imposed a custodial sentence of one year. The respondent submitted in oral argument, and my colleague McLachlin J. appears to have accepted this submission, that A.B.C. was decided on the basis of procedural delays and other particular facts and should not affect the present analysis. On the contrary, to the extent that the particular facts in A.B.C. determined the sentence in that case, it may also be equally argued that the particular facts of the present case determined the one-year sentence meted out by the sentencing judge for a sexual assault similar in nature to that in A.B.C. It is difficult to conclude that a sentence of one year in the present case, given the similarities to A.B.C., was demonstrably unfit. While this sentence is at the bottom of the scale, this does not make it demonstrably unfit. I note that in both Shropshire and M. (C.A.), two recent, unanimous decisions, this Court refused on the basis of deference to reduce sentences that were clearly at the high end of the spectrum.

26In summary, with respect to the first assault, the trial judge did not fail to consider relevant factors, nor was the sentence demonstrably unfit.

27With respect to the second offence, the same conclusion applies: the judge did not ignore the factors raised by the Court of Appeal, nor is there any indication that the sentence was demonstrably unfit. The sentencing judge did, as the Court of Appeal acknowledged (at p. 176), consider the trauma to the complainant, stating:

Regarding the charges of assault against [the second complainant], the victim in this case has been traumatized, but the acts of the accused were very much in the ‘less grave’ category. Mr. McDonnell is a man of otherwise good character and is a strong member of his community. He has always maintained employment and supported his family. An additional lengthy consecutive custodial sentence to the custodial sentence imposed on the first charge would only seek to destroy the accused and his family and is not necessary to deter others from committing such an offence.

I will thus sentence him to six months in gaol concurrent to the first sentence, plus probation for the same period of time.

Neither the Court of Appeal nor the respondent points out a factor ignored by the sentencing judge in reaching her conclusion of a sentence of six months for the second offence. Nor is there any reason given by either the respondent or the Court of Appeal to conclude that a six-month sentence was demonstrably unfit. The Court of Appeal simply disagreed with the sentence ordered and substituted its own opinion for that of the sentencing judge.

28My colleague McLachlin J. disagrees with this analysis and states that the sentences imposed by the sentencing judge were outside the acceptable range. While the above analysis generally addresses her reasoning, I add here that I disagree with her conclusion that a variety of past cases reveals that the sentence in the present case was demonstrably unfit. McLachlin J. provides at para. 110 a lengthy list of cases which she contends support the conclusion that a sentence under two years in the present case was inappropriate. While I will not review each case upon which she relies, I will note that, in my view, many of the cases provided are inappropriate cases to consider in the present context. For example, in R. v. S.G.O.R. (1991), 113 A.R. 36 (C.A.), aside from other sexual offences involved in the case, the accused raped his daughter over 20 times from when she was four years old until she was 12. In R. v. S. (W.B.); R. v. P. (M.) (1992), 1992 CanLII 2761 (AB CA), 73 C.C.C. (3d) 530 (Alta. C.A.), the accused S. engaged repeatedly in anal intercourse with his six-year-old stepdaughter and his stepson, who was initially in grade three, over a period of two years. The accused P. committed both anal and vaginal rape of a seven-year-old child, physically beating her head and body. R. v. Spence (1992), 1992 ABCA 352, 78 C.C.C. (3d) 451 (Alta. C.A.), involved an accused raping his 15-year-old cousin and physically beating and threatening her. R. v. Nicholson (1993), 1993 ABCA 248, 145 A.R. 262 (C.A.), involved 20 to 30 acts of intercourse starting when the complainant was 12 years old; associated with this abuse, the accused, amongst other things, discharged a rifle in the direction of the complainant as she attempted to escape his residence. Other cases cited by McLachlin J. involved children much younger than the complainants in the present case, such as R. v. Lapatak (1995), 1995 ABCA 322, 169 A.R. 385 (C.A.), which involved a three-year-old victim.

29In my view, many of the cases cited by McLachlin J. involved offences considerably more serious than the present case. While any sexual offence is serious, particularly on young people, the violence of the offences, the repetition of the offences, and the extreme youth of the victims in the cases cited above clearly distinguish them from the present case. Indeed, contrary to supporting McLachlin J.’s position, in my view the variable circumstances in the cases she cites highlight the importance of individualized sentencing. In any event, in my respectful view, the sentences in the cases she cites do not lead to the conclusion that the sentences in the present case were demonstrably unfit. The sentences in the present case, while low, were not demonstrably unfit.

C. Error in Principle

30While I have concluded that the sentencing judge did not ignore factors and that the sentences were not demonstrably unfit, according to M. (C.A.) and Shropshire, appellate review of a sentence is also appropriate if the sentencing judge committed an error in principle. Both the Court of Appeal and the respondent appear to treat the alleged departure from Sandercock as an error in principle. For example, the Court of Appeal found and the respondent submitted that the first assault was a “major sexual assault” in contradiction to the finding of the sentencing judge. The Court of Appeal stated (at pp. 172-73):

We have concluded that the trial judge erred in finding that this was not a major sexual assault. This court made it clear in R. v. Sandercock ..., that the key to a major sexual assault is the “evident blameworthiness of the offender” as reflected in the extent to which the offender’s actions demonstrated a “contemptuous disregard for the feelings and personal integrity of the victim”. Here, McDonnell’s actions in the first case clearly fall within the category of a major sexual assault.

The court then treated the error as one which justified alteration of the sentence, which implicitly treated the failure to find the major sexual assault as an error in principle. The Court of Appeal concluded that the sentencing judge wrongly declined to find a major sexual assault in part on the basis of what the court viewed as a misapprehension of the requirements for a major sexual assault. The court stated that the sentencing judge appeared to accept the argument of the defence that there was no significant psychological harm in the first case, and therefore there was no major sexual assault. The court stated (at p. 173) that while actual psychological harm may generally be presumed from non-consensual intercourse, it is not the actual harm, but the high likelihood of harm from the nature of the assault that gives rise to a major sexual assault.

31I disagree with the Court of Appeal that the sentencing judge accepted the argument of the defence that there was no psychological harm. As noted above, the sentencing judge specifically found psychological harm to the complainants in both cases. In my view, even if the sentencing judge required proof of such harm before finding a major sexual assault, the sentencing judge appeared to find such harm so any “error” in this regard did not affect the outcome.

32In any event, in my view it can never be an error in principle in itself to fail to place a particular offence within a judicially created category of assault for the purposes of sentencing. There are two main reasons for this conclusion. First, Shropshire and M. (C.A.), two recent and unanimous decisions of this Court, clearly indicate that deference should be shown to a lower court’s sentencing decision. If an appellate court could simply create reviewable principles by creating categories of offences, deference is diminished in a manner that is inconsistent with Shropshire and M. (C.A.). In order to circumvent deference and to enable appellate review of a particular sentence, a court may simply create a category of offence and a “starting point” for that offence, and treat as an error in principle any deviation in sentencing from the category so created. Indeed, that is what the Court of Appeal in Alberta has done in the present case. If the categories are defined narrowly, and deviations from the categorization are generally reversed, the discretion that should be left in the hands of the trial and sentencing judges is shifted considerably to the appellate courts.

33Second, there is no legal basis for the judicial creation of a category of offence within a statutory offence for the purposes of sentencing. As has been true since Frey v. Fedoruk, 1950 CanLII 24 (SCC), [1950] S.C.R. 517, it is not for judges to create criminal offences, but rather for the legislature to enact such offences. By creating a species of sexual assault known as a “major sexual assault”, and by basing sentencing decisions on such a categorization, the Alberta Court of Appeal has effectively created an offence, at least for the purposes of sentencing, contrary to the spirit if not the letter of Frey.

34The danger of courts encroaching into the realm of Parliament by creating offences is illustrated by the present case. The Court of Appeal appeared to base its conclusion that the first assault was a “major sexual assault” on the likelihood of psychological harm and indeed on the existence of actual harm. The court thus concluded that the sentence should be based on the existence of such harm. There is, however, a specific offence that deals with sexual assault causing bodily harm within the Criminal Code, namely s. 272(c). I note that R. v. McCraw, 1991 CanLII 29 (SCC), [1991] 3 S.C.R. 72, established that psychological harm from a sexual assault may be considered bodily harm. Given Parliament’s intention to treat sexual assaults causing bodily harm under s. 272(c), it is particularly inappropriate to create a “major sexual assault”, which is based at least in part on the existence of harm to the complainant pursuant to s. 271. While the Court of Appeal at times appeared to rely simply on the likelihood of harm in the present case, in Sandercock itself, actual harm was contemplated. Sandercock stated at p. 85, “[t]he other aspect which creates a major sexual assault is the effect on the victim.” In my view, if the prosecution is to be based on the harm to the victim, the accused should be charged under the appropriate section, s. 272(c). It is not for the courts to establish a subset of offence within s. 271 that is based on harm.

35There is a further problem with the treatment of harm by the Court of Appeal in the present case in that it appeared at times to establish a presumption of psychological harm from a sexual assault. Admittedly, at other times the Court discussed the likelihood of psychological harm from an offence as illustrating the seriousness of the offence, rather than actual harm itself. To illustrate this ambiguity, consider the following passage (at p. 173):

The first point we wish to make is that we cannot envision a situation where nonconsensual intercourse vaginal, anal or oral would not fall into the major sexual assault category.... In addition, in each case, there also exists a very real likelihood of psychological harm. Therefore, what must be understood is that it is not necessary that the Crown prove the existence of this kind of harm as a condition precedent to the courts classifying a sexual assault as a major one. Psychological harm is presumed in the absence of evidence to the contrary. [Emphasis added.]

The court later stated (at p. 174):

To put the matter another way, the offender is being sentenced on the basis of a major sexual assault, not because any specific psychological consequences have flowed from the attack but rather because of the nature of the attack and the fact that it poses the very real likelihood of long-term emotional or psychological harm. The fact that no such harm may materialize, a fact one could not possibly know until the victim's life had been lived in its entirety, is not a mitigating factor. However, that said, this does not mean that the consequences of the sexual assault are irrelevant. The degree of seriousness of the actions may be measured against the likely long-term consequences of the prohibited act. In other words, where the psychological harm has been severe, that may well be an aggravating factor. Of course, where harm beyond that which would be normally presumed is claimed in a case, the Crown must lead evidence to substantiate it. [Emphasis in original.]

36These passages are somewhat unclear. At one point it appears that the court presumes that psychological harm would result from a sexual assault, while at another point it appears that the court is not presuming psychological harm, but rather is simply noting, correctly in my view, the likelihood of psychological harm resulting from the actions of the accused. McCraw, supra, established that a threat to commit sexual assault amounted to a threat to commit assault causing bodily harm because of the high likelihood of psychological harm resulting from a sexual assault, a likelihood recognized by the Court of Appeal in the present case. Such a likelihood does not, however, establish a legal presumption of harm in cases involving an actual assault, as opposed to a threat. If harm is an element of the offence, the Crown must prove its existence beyond a reasonable doubt.

37To the extent that the Court of Appeal held that the Crown need not prove psychological harm in some instances, but rather such harm may be presumed, it was in error. As stated above, if the Crown wishes to rely upon the existence of psychological harm, in my view the Crown should charge under the section set out in the Code that contemplates harm, s. 272(c), and prove the offence. If an element of the offence, bodily (psychological) harm, is presumed, the Crown is improperly relieved of part of the burden of proof, which is contrary to the presumption of innocence. Accepting that harm may be an aggravating factor under s. 271, R. v. Gardiner, 1982 CanLII 30 (SCC), [1982] 2 S.C.R. 368, held that each aggravating factor in a sentencing hearing must be proved beyond a reasonable doubt. Such an approach is confirmed by Parliament in the new s. 724(3)(e) of the Criminal Code (as amended by S.C. 1995, c. 22, s. 6). If psychological harm may be presumed, the burden of proving harm as an aggravating factor is improperly lifted from the Crown and shifted to the accused to disprove harm.

38In the present case, a presumption of harm is unnecessary. The sentencing judge found as a fact that each complainant in the present case was traumatized. The sentence was reached after considering the harm that resulted from the offences. Thus, the Court of Appeal’s discussion of the presumption of harm was, in my view, both erroneous and unnecessary; harm existed and was considered in setting the sentence.

39The Court of Appeal appeared to make two other suggestions of errors in principle by the trial judge. I note that the respondent did not specifically raise these alleged errors in written argument before this Court, but raised them specifically only in oral argument. One error alleged by the Court of Appeal was that the trial judge improperly relied on R. v. R.P.T. (1983), 1983 ABCA 175, 7 C.C.C. (3d) 109 (Alta. C.A.). The court stated that, notwithstanding that the principles it set out were revisited in R. v. S. (W.B.); R. v. P. (M.), supra, R.P.T. was factually inapplicable because in the present case there was no family that might be restored. In my view, the court erred in finding that the trial judge relied on R.P.T. with respect to the restoration of the family. R.P.T. held at p. 114 that even where there is a family to restore, if the sexual assault by a person in loco parentis on a family member were serious, “[t]he only solution, however imperfect ... must be to graft a rehabilitative sentence to a denunciatory sentence”. If the assault were less serious and if there were a family to restore, a lesser sentence may be imposed; indeed, if the circumstances were “less significant” (p. 115), a suspended sentence may be appropriate.

40In the present case, the sentencing judge stated that while she was cognizant of R.P.T., “this is not a case where simple rehabilitation will suffice”. She stated that a rehabilitative sentence must be grafted to a denunciatory sentence. She thus apparently relied on the aspect of R.P.T. which held that despite the existence of a family, both a denunciatory and a rehabilitative sentence are required where the assault is serious. Contrary to the position of the Court of Appeal, she did not rely on the aspect of R.P.T. which stated that restoration of the family should be a mitigating consideration in cases where the assault is less serious. Given that the sentencing judge did not follow R.P.T to rely on family restoration to mitigate the sentence, the Court of Appeal failed to point out an error in principle by the sentencing judge with respect to R.P.T.

41Another error suggested by the Court of Appeal was that the sentencing judge improperly relied upon the passage of time “between the first and second offence (and sentencing)” (p. 177). The sentencing judge did not in any way rely upon the passage of time between the first offence and the second offence. Neither did the sentencing judge rely upon the passage of time between the first offence and sentencing per se, but rather the judge stated that “the time elapsed since the offence has relevance in relation to the relative effect of this on both the accused and the victim”. I presume that the sentencing judge was referring to effect of the actions on the accused since the offence, which included, for example, remorse and a desire to quit drinking, and to the effect of the actions on the victim since the offence, which included the psychological harm that the victim had displayed since the offence. These factors may be relevant considerations and the sentencing judge did not err in principle in referring to them.

42I note that my colleague McLachlin J. states that she agrees that failing to characterize the offence into a particular, judicially created category of assault is not an error in principle which would justify appellate review. However, I am concerned that while she states she does not view it as such an error, she effectively treats it, if not as an error in principle, then otherwise as an error giving rise to appellate review. She states at para. 109:

As indicated earlier, the “starting point” is not a principle of law, but rather a tool to determine the proper range of sentence for a certain type of offence. Failure to allude to the appropriate starting point or range is not an error of principle as that term is used in M. (C.A.), supra. If the trial judge fails to refer to the appropriate starting point or range but in the end imposes a sentence within the acceptable range of sentence for the offence as adjusted for the particular circumstances of the offender, a court of appeal should not interfere. On the other hand, if the sentence falls outside the appropriate range, the court of appeal must interfere: Shropshire, supra.

This statement, combined with her emphasis on starting points in her analysis of demonstrable unfitness in the present case, suggests to me that McLachlin J. in effect treats the failure to characterize an assault properly as an error permitting appellate intervention on sentencing. That is, the failure to characterize the assault properly is not an error in principle, but if the sentence reached as a result of that error is not very similar or identical to the sentence that would have been reached had the mischaracterization not occurred, appellate courts may intervene. In my view, this effectively states that while appellate courts must permit sentencing judges to err in characterizing the offence, appellate courts may intervene, notwithstanding deference, if the trial judge’s mischaracterization affected significantly the sentence ordered. Given that different views of the nature of the assault would almost inevitably lead to different sentences, in my view, mischaracterization is treated by McLachlin J. as an error which will often lead to appellate intervention. In my view, as stated, mischaracterization of the offence according to judicially created categories is not an error in principle, nor should it be treated as one. In my respectful opinion, McLachlin J. takes an overly permissive approach to appellate intervention that is inconsistent with both Shropshire and M. (C.A.).

43I add that I do not disagree with McLachlin J. that appellate courts may set out starting-point sentences as guides to lower courts. Moreover, the starting point may well be a factor to consider in determining whether a sentence is demonstrably unfit. If there is a wide disparity between the starting point for the offence and the sentence imposed, then, assuming that the Court of Appeal has set a reasonable starting point, the starting point certainly suggests, but is not determinative of, unfitness. In my view, however, the approach taken by McLachlin J. in the present case places too great an emphasis on the effect of deviation from the starting point. Unless there otherwise is a reason under Shropshire or M. (C.A.) to interfere with the sentence, a sentence cannot be altered on appeal, notwithstanding deviation from a starting point. Deviation from a starting point may be a factor in considering demonstrable unfitness, but does not have the significance McLachlin J. gives it.

D. Concurrent or Consecutive Sentences

44The Court of Appeal not only substituted its view of the appropriate sentence for each offence in the present case, but also held that the sentences should be served consecutively, not concurrently as had been ordered by the sentencing judge. The Court of Appeal stated (at p. 177):

Nor was this a case where concurrent sentences were appropriate.... These two offences were totally unrelated. There were two different victims and the offences were separated by a period of seven years.

45The sentencing judge considered whether to order consecutive and concurrent sentences and concluded that concurrent sentences were appropriate, stating:

An additional lengthy consecutive custodial sentence to the custodial sentence imposed on the first charge would only seek to destroy the accused and his family and is not necessary to deter others from committing such an offence.

46In my opinion, the decision to order concurrent or consecutive sentences should be treated with the same deference owed by appellate courts to sentencing judges concerning the length of sentences ordered. The rationale for deference with respect to the length of sentence, clearly stated in both Shropshire and M. (C.A.), applies equally to the decision to order concurrent or consecutive sentences. In both setting duration and the type of sentence, the sentencing judge exercises his or her discretion based on his or her first-hand knowledge of the case; it is not for an appellate court to intervene absent an error in principle, unless the sentencing judge ignored factors or imposed a sentence which, considered in its entirety, is demonstrably unfit. The Court of Appeal in the present case failed to raise a legitimate reason to alter the order of concurrent sentences made by the sentencing judge; the court simply disagreed with the result of the sentencing judge’s exercise of discretion, which is insufficient to interfere.

V. Conclusion and Disposition

47The appellant raised a variety of issues in this appeal aside from those discussed in these reasons. Given my analysis herein, it is unnecessary to consider these other grounds. In my view, the appeal should be allowed and the sentences ordered by the sentencing judge restored.

The reasons of La Forest, L’Heureux-Dubé, Gonthier and McLachlin JJ. were delivered by

48McLachlin J. (dissenting) -- A just sentence is one which reflects the seriousness of the crime and fits the individual circumstances of the offender. Appellate courts in a number of provinces have responded to the need to accommodate these dual goals and resolve the tension which may arise between them by introducing the concept of “starting-point” sentences. This appeal raises the question of whether the “starting-point” approach to sentencing offends the guarantees of fundamental procedural justice enshrined in the Canadian Charter of Rights and Freedoms.

49It is my view that starting-point sentencing, properly understood and applied, does not violate the Charter. Rather, it represents a fair and practical way at arriving at a just sentence that respects both the individual circumstances of the offender and the seriousness of the offence. In my view, the Court of Appeal did not err in applying this approach. I would thus dismiss the appeal.

I. The Facts

50The appellant pleaded guilty to two charges of sexual assault. The first assault, in 1986, was on a girl living in the appellant’s home as a foster child. The appellant was 29, the girl 16. The incident occurred about two weeks after the girl came to live with the family. The appellant arrived home intoxicated and found the victim asleep on her back on the living room couch. He undid her jeans. She rolled onto her stomach and tried to press herself into the couch in hope that he would go away. He did not. Instead, he removed her jeans and began to kiss her buttock area. He penetrated her vagina with his penis. She had her legs pressed together and he was trying to get them apart. Nevertheless, he penetrated “a little bit”. Finally, the appellant said, “You’re too difficult”, and rolled onto the floor. The girl waited until she was sure he was asleep and went downstairs to her room. At no time did she consent to his acts.

51The second assault occurred in 1993 when the appellant was 36 years of age. The victim was a 14-year-old babysitter for the appellant’s family. While babysitting, she had fallen asleep on her stomach on a hide-a-bed. She was wearing a T-shirt and underwear and was wrapped with a sleeping bag. She woke at 3:30 a.m. to find her underwear pulled down and the sleeping bag around her feet. The appellant was on top of her rubbing her back under her T-shirt with one hand and her buttocks with the other. He reached under her stomach and touched her pelvis and vaginal areas. He tried to turn her over onto her stomach. She screamed and fled the house.

52The evidence showed that both victims suffered psychological harm as a result of the assaults. The first victim testified on the preliminary inquiry that the act “hurt me mentally”. The victim impact statement in the sentencing material attested to the fact that the sexual assault harmed her. As for the second victim, the trial judge found that she was, to use her words, “traumatized” by the assault.

II. The Rulings

53The trial judge did not consider either sexual assault to fall within the category of a major sexual assault calling for a “starting point” of three years’ imprisonment. As the trial judge did not clearly differentiate between considerations going to whether the sexual assault was major and considerations individualizing the sentence, it is difficult to determine what she took into account in deciding that neither assault constituted a major sexual assault. However, in stating that the first assault was not major, she mentioned the fact that it was an isolated one, that the penetration was only partial, that no violence had been used, that there had been no oral sex or anal intercourse, that the event was “spontaneous due to [the appellant’s] drunkenness”, and that while it was a traumatic experience for the victim, she was having other problems which may have contributed to her state of mind. Of the second assault, she stated merely that while the victim had been “traumatized”, the “acts of the accused were very much in the ‘less grave’ category”. She went on to discuss the appellant’s good character and the destructive effect of a sentence of long incarceration on him and his family.

54Having concluded that neither assault was major, the trial judge acknowledged the need for a sentence reflecting denunciation and the goal of general deterrence. She stated that the sentence must consider more than simple rehabilitation. On the other hand, she noted the appellant’s desire to stop drinking, his strong family and community support, the fact that the first victim’s trauma was suffered when she was already troubled and the fact that counselling would be of no use to prevent re-offending. These factors and the effect long-term incarceration would have on the accused and his family led her to impose a sentence of 12 months’ incarceration for the first offence and six months for the second, to be served concurrently.

55The Alberta Court of Appeal concluded the trial judge had erred in not considering the assaults major and replaced the 12-month-plus probation sentence with a sentence of four years for the first assault, followed by an additional year for the second assault, for a total of five years’ imprisonment: (1995), 1995 ABCA 200, 169 A.R. 170, 97 W.A.C. 170. In the Court of Appeal’s view, the first assault was an “archetypical” example of a major sexual assault, making a “starting point” of three years applicable. The appellant had breached the trust reposed in him as a foster parent. He had done so in a manner which involved a serious violation of the victim’s sexual and personal integrity and demonstrated a complete lack of concern for the feelings of the young girl. The assault involved penile penetration. In the case of such an assault, harm can be presumed or inferred, even in the absence of evidence. But even if that were not so, here there was evidence of trauma. Severe trauma is not required to bring the case within the category of a “major sexual assault”, although it might serve as an aggravating factor. The passage of time, while relevant, did not assist the appellant. It was not lengthy and, far from living a blameless life in the interval, the appellant had re-offended. Since the offences were unrelated, concurrent sentences were not appropriate. The Court imposed a sentence of four years for the first offence, followed by a sentence of one year for the second, taking into account the global effect of the sentences and the lesser severity of the second assault.

III. The Issue

56Did the Court of Appeal err in overturning the trial judge’s one-year sentence and replacing it with a cumulative sentence of five years for the two offences? The appellant, in support of his submission that it did, argues that the starting-point approach used by the court violates his rights, that the Court of Appeal in effect created a new offence known as a major sexual assault, that the Court of Appeal erred in presuming psychological harm, and finally, that the Court of Appeal improperly interfered with the trial judge’s discretion and impermissibly substituted its own view of the matter.

IV. Analysis

A. The Starting-Point Approach to Sentencing

(1) What Is the Starting-Point Approach to Sentencing?

57My difficulty with the position of the appellant and the reasons of Sopinka J. stems mainly from a different understanding of the nature and effect of the “starting-point” approach to sentencing. It is therefore necessary to set out my conception of that approach at the outset.

58The starting-point approach to sentencing involves two steps. First, the judge determines the range of sentence for a typical case. Using that range as a starting point, a trial judge then adjusts the sentence upward or downward on the basis of factors relating to the particular offence and offender: R. v. Hessam (1983), 1983 ABCA 69, 43 A.R. 247 (C.A.), R. v. Sandercock (1985), 1985 ABCA 218, 22 C.C.C. (3d) 79 (Alta. C.A.). This approach is distinguished from the tariff approach to sentencing which takes no account of the individual circumstances of the offender: C. C. Ruby, Sentencing (4th ed. 1994), at p. 479. The tariff approach looks only at the nature of the offence. In contrast, the starting-point approach mandates consideration of specific aggravating and mitigating factors directly relevant to the individual accused. In this way, the starting-point approach combines general considerations relating to the crime committed with personalized considerations relating to the particular offender and the unique circumstances of the assault.

59The first step on the starting-point approach consists of determining the appropriate range of sentence for an offence of this type in a typical case, assuming an offender of good character with no criminal record. In the case of sexual assault, the judge looks at the manner in which the assault was committed (e.g., by violence or threats or trickery), the nature of the sexual activity, and, most importantly, whether or not this sort of offence is likely to cause lasting emotional or psychological injury. The “key ... to a major sexual assault is the evident blameworthiness of the offender”, the “contemptuous disregard for the feelings and personal integrity of the victim”: Sandercock, supra, at p. 84. The inquiry at this stage, to repeat, is generalized and objective. The task of the judge at this stage is to determine the blameworthiness of an offender who commits the type of offence at issue in a typical case. Because the inquiry at this point is general, it proceeds on certain assumptions. The issue, on harm, is not whether actual trauma occurred, but whether this sort of criminal act would be likely to cause lasting emotional or psychological trauma. As to the offender, it is assumed that the offender is of good character and has no criminal record. See Sandercock, supra.

60The exercise of choosing a starting point in this way resembles the long-standing practice of setting a range of sentence as a tool to arrive at a just and appropriate sentence that reflects both the crime and the individual circumstances of the offence and the offender. As Ruby, supra, at p. 482, notes, “[i]t certainly is not a new method of sentencing”. The starting point may be viewed as the mid-point in the traditional range of sentences for a particular sort of crime.

61The choice of a starting point is only -- as the phrase makes clear -- a starting point. Based as it is on assumptions as to the harm likely to flow from a typical case of the type of criminal act and the good character of the accused, it could not in fairness or principle serve as a final indication of the appropriate sentence in a particular case. As noted in Sandercock, supra, every case has its own unique characteristics, and every offender his or her own unique history. The goals of sentencing -- deterrence, retribution and rehabilitation -- play out differently depending on the peculiar concatenation of circumstances presented in each case. In short, the sentence must be individualized to the particular crime and the particular offender before the court. Having determined a starting point, the judge must go on to consider these factors and their effect on the appropriate sentence. The factors peculiar to the particular case and offender before the court may mitigate, resulting in a lower sentence than the typical case reflected by the starting point. Or they may exacerbate, resulting in a higher sentence than would prevail in the typical case.

(2) Why Was the Starting-Point Approach Developed?

62The starting-point approach was developed as a way of incorporating into the sentencing process the dual perspectives of the seriousness of the offence and the need to consider the individual circumstances of the offender. It represents a restatement of the long-standing practice of sentencing judges of beginning by considering the range of sentence that has been posed for similar criminal acts followed by consideration of factors peculiar to the case and offender before them.

63Despite the common practice of first determining a range and then individualizing the sentence, the jurisprudence dealing with the proper approach to sentencing is not as clear as might be desired. Professor A. Young, The Role of an Appellate Court in Developing Sentencing Guidelines (1988), a report written for the Canadian Sentencing Commission, offers a useful history of sentencing theory in Canada and the failure of the courts to adequately meet the challenge of devising a principled and consistent approach to sentencing.

64Appellate review of sentences was initiated only in 1921, explaining the absence of long-standing principles to guide trial judges. Prior to 1921, trial judges gave the sentence they saw fit and that was the end of the matter. Nor, in the years after 1921, were the courts instrumental “in designing relevant sentencing principles to assist lower-courts .... Only in recent years have the appellate courts begun to express dissatisfaction with the impressionistic nature of sentencing decisions” (Young, supra, at p. 6). The maxim “Let the punishment fit the crime” might rule on Gilbert and Sullivan’s stage, but in the courts the theme was “that the punishment should fit the offender” (Young, at p. 8). Precedents and theory played little part in the sentencing process. “In a sentencing model based upon the primacy of the individual there is little need for precedents that can extend beyond the characteristics of the offender in any given case” (Young, at p. 8). The Saskatchewan Court of Appeal (R. v. Natanson (1927), 1927 CanLII 518 (SK CA), 49 C.C.C. 89) put the conventional wisdom this way (at p. 90):

It would be impossible, and if possible it would be undesirable to lay down any general rule as to the punishment to be inflicted for any particular class of offence. Every case must be dealt with on its own facts and circumstance[s].

Similarly, in R. v. Connor and Hall (1957), 1957 CanLII 509 (ON CA), 118 C.C.C. 237, the Ontario Court of Appeal opined (at p. 238):

It serves little useful purpose and affords little assistance to the Court to know what sentences have been imposed in other countries or jurisdictions or by other Courts.

65The traditional notion that sentencing is primarily a matter of impression for the sentencing judge and only secondarily a matter of principle began to be questioned by the courts in the mid-60s. Behind the challenge lay increasing recognition that some measure of uniformity was essential in a sentencing process that not only was just, but was perceived to be just. In R. v. Baldhead, 1966 CanLII 335 (SK CA), [1966] 4 C.C.C. 183 (Sask. C.A.), it was held that a sentence could be reviewed if it represented “a marked departure from the sentences customarily imposed in the same jurisdiction for the same or similar crimes” (p. 187).

66Baldhead did no more than confer judicial respectability on an emerging general consensus that the law should award similar sentences for similar crimes, subject to adjustment for factors peculiar to each case. Sentences may properly vary somewhat from case to case to reflect factors peculiar to the particular act and offender on trial. But it affronts common-sense notions of justice if people who have committed the same criminal act receive wildly disparate sentences. It is neither fair nor just that one person languish in prison years after another, who committed a similar act, is released to liberty. Baldhead expressed the growing view that a measure of uniformity, tempered but not obliterated by considerations particular to each case, must stand as a fundamental goal of sentencing law.

67Many courts since Baldhead have embraced the objective of uniformity as a factor to be considered in sentencing. However, the relationship between the goal of sentencing uniformity and the goal of reflecting in a sentence the circumstances of the particular case and offender remained largely ill-defined up until the jurisprudence advocating a starting-point approach. Alongside decisions advocating the need for a measure of uniformity, stand other decisions evincing reluctance to commit it to principle. As Young, supra, puts it: “The courts have not wholly embraced the notion of uniformity for fear that broad, general principles will fail to take into account the unique characteristics of every offender” (pp. 9-10). By contrast, the starting-point approach represents an attempt to marry in one sentencing principle the values of uniformity and individualization.

68It was no accident that the starting-point approach was eventually applied in the context of the crime of sexual assault. The wide spectrum of conduct embraced by the crime of “sexual assault” and the disparate views different judges may take with respect to the gravity of particular types of sexual assaults give rise to wide variations in sentences for offences that seem quite similar. See P. Marshall, “Sexual Assault, The Charter and Sentencing Reform” (1988), 63 C.R. (3d) 216. Depending on where a particular judge placed a particular type of sexual assault on the spectrum of severity and the seriousness with which he or she regarded that assault, a sentence might be high or low or anywhere in between. The disparities between sentences threatened to go beyond the legitimate area of divergence represented by the individual circumstances of a particular offender and offence, to a more generalized divergence based on judicial views of the seriousness of the offence. This called for judicial action. As the Manitoba Court of Appeal put it in R. v. Jourdain and Kudyba (1958), 1958 CanLII 445 (MB CA), 121 C.C.C. 82, at p. 87:

It is the duty not only of this Court but of all the Courts of the Province and the Crown to do whatever is possible to bring about uniformity and equalization of sentences for crimes of the same or similar gravity.

The response of courts, charged as they were with maintaining reasonable uniformity of sentences, was to introduce the concept of the “starting point”.

(3) In What Jurisdictions Has the Starting-Point Approach Been Adopted?

69The Courts of Appeal for Alberta, Nova Scotia (R. v. Zong (1986), 1986 CanLII 6903 (NS CA), 173 A.P.R. 432), Manitoba (R. v. Muswagon (1993), 1993 CanLII 14904 (MB CA), 88 Man. R. (2d) 319), British Columbia (R. v. Post (1996), 72 B.C.A.C. 312) and Saskatchewan (R. v. Jackson (1993), 1993 CanLII 4414 (SK CA), 87 C.C.C. (3d) 56) have applied the starting-point approach to sentencing to deal with marked disparities in sentences for certain crimes. The Ontario Court of Appeal in R. v. Glassford (1988), 1988 CanLII 7094 (ON CA), 27 O.A.C. 194 explicitly rejected the starting-point approach to sentencing articulated in Sandercock, supra. However, that same court has recently adopted the approach in narcotics cases. See R. v. Cunningham (1996), 1996 CanLII 1311 (ON CA), 104 C.C.C. (3d) 542.

70In addition to the Canadian examples, the English Court of Appeal also appears to have adopted this approach. In R. v. Edwards; R. v. Brandy, the English Court of Appeal, Criminal Division, suggested that “[a]n appropriate level of sentencing for serious dwelling house burglary where the house was unoccupied was three years on a conviction, with variations either way to reflect the particular circumstances of the case” (The Times, July 1, 1996). In fact, it appears that the starting-point concept is not of recent origin in England. Cross, The English Sentencing System (2nd ed. 1975), states at p. 148:

The statement of 1900 is contained in a Memorandum produced by Lord Alveston (the then Lord Chief Justice) in an effort to get agreement among the Queen’s Bench Judges about the normal punishment of offences. It was sent to the Home Office and no further action appears to have been taken on it; but it is now printed as Appendix 5 of Enforcing the Law by Professor Jackson of Cambridge. The Memorandum states that it is not possible to do more than recommend “a range of punishments within certain limits” and throughout it speaks of periods such as three to five years penal servitude as the “correct range”. When dealing with rape, for instance, the Memorandum mentions five to seven years penal servitude as giving:

“a reasonable range of punishment to be increased if there are accompanying circumstances of aggravation, such, for example, as rape by a gang or by a parent or master, or with brutal violence, and to be reduced if there are extenuating circumstances.”

71The Australian courts appear to follow a similar approach. In R. v. Jabaltjari (1989), 46 A. Crim. R. 47, the Court of Criminal Appeal (Northern Territory) did not interfere with the trial judge’s approach described as follows: “Having fixed on the objective sentence his Honour then made appropriate adjustments downward to give effect to the mitigating circumstances personal to the respondent” (p. 64). Although the approach is referred to by the Court as the “tariff” approach, it seems to be identical to the starting-point approach.

(4) Academic Opinion

72Academic comment appears to favour the starting-point approach. Ruby, supra, at pp. 477-78, writes:

Uniformity, to the extent that it is desirable, means uniformity in approach. Acceptance of this principle will minimize the number of cases where wide disparity in sentence exists.

This necessarily implies that “where there is a marked departure from the sentences customarily imposed in the same jurisdiction ... the appellate Court ... should be able to rationalize the reason for departure.”

73Having noted the need for a measure of uniformity in sentencing, Ruby expressly endorses the move to starting points for sentencing (at pp. 481-82):

[The approach] may have autocratic aspects, one suspects, but the approach may be helpful. It certainly is not a new method of sentencing, and it is quite likely to produce a proper sentence. We are seeing no more than a court deciding to take seriously the accepted notion that what is required is a uniform approach to sentencing.

74Others who have studied sentencing likewise conclude that it is desirable that the courts, and if necessary Parliament, develop principles to promote greater uniformity in sentencing. The 1987 Report of the Canadian Sentencing Commission, Sentencing Reform: A Canadian Approach, went beyond guidelines imposed by courts of appeal to recommend legislated sentencing guidelines. Such guidelines, the Commission was at pains to point out, must be “distinguished from a form of mandatory prescription”. Judges can and should depart from even legislated guidelines in appropriate cases. See Chapter 11, at pp. 271-72. The Commission’s recommendation for guidelines has been welcomed: see Marshall, supra, at p. 222.

75Young, supra, likewise advocates the development of doctrine to ensure a greater measure of uniformity and consistency in sentencing. Referring to judicial attempts to do so, he writes (at p. 98):

Efforts have been made to introduce these various measures; however, their introduction has been impeded by a lack of commitment to the importance of the enterprise. The discussion presented in this paper clearly shows that the appellate courts have been irresolute in their efforts. Perhaps the appellate courts have been discouraged by the recognition that their modest efforts have been largely disregarded by the lower courts. For whatever reason, we are left with a process that cannot answer with any certainty why it is that one offender receives a non-custodial term and another similarly-situated offender is subject to a term of imprisonment.

Young goes on (at pp. 98-99):

There is general agreement that the criminal justice system has failed to remedy manifest inequities in the sentencing process .... [A] solution is not beyond the institutional competence of the judiciary; they possess the requisite expertise and familiarity with the issues to enable them to develop coherent and consistent guidelines.

76A. M. Linden, in “A Fresh Approach to Sentencing in Canada”, in H. Dumont, ed., Sentencing (1987), 53, writes at p. 56:

The most serious problem with our current system is the disparity in treatment it actively promotes. Substantially similar crimes may give rise to widely, and wildy [sic], differing sentences. This disparity, or lack of equality, commands particular attention with the coming into force of the equality rights provisions of the Canadian Charter of Rights and Freedoms.

Linden then goes on to advocate in favour of a guidelines approach to sentencing that involves setting benchmarks which can vary depending on aggravating or mitigating factors.

77The development of the starting-point approach at issue on this appeal may be seen as an attempt to resolve the tension between individual sentencing and the need for greater uniformity and consistency in sentencing in one area of the criminal law, the offence of sexual assault.

(5) The Arguments Against a Starting-Point Approach

78The starting-point approach appears to meet both the requirements of uniformity and consistency in sentencing and individualized justice. It presents itself as a positive step toward a more rational approach to sentencing which the public and commentators alike have called for. Nevertheless, the appellant argues that this Court should reject it, a view accepted by Sopinka J. I turn then to the arguments against the starting-point approach.

(a) The Argument that the Starting-Point Approach Fetters Judicial Discretion to Impose Individualized Sentences

79It is argued that the starting-point approach to sentencing fetters the sentencing judge’s discretion and is thus contrary to the fundamental principle of justice that sentences must be individualized to the circumstances of the offender. In my view, the starting-point approach facilitates, rather than hinders, the proper exercise of judicial discretion and the individualization of sentences. The starting point is only the beginning of the sentencing exercise. Having fixed the starting point, the judge is obliged to consider aggravating or mitigating factors. As stated in Sandercock, supra, the approach “does not arbitrarily confine the discretion of the sentencing court. Rather, it offers a rational structure for its exercise, and a structure which is just because it guards against both disparity and inflexibility” (p. 82).

80A properly chosen starting point does not fetter discretion, but confines it to legitimate considerations. The fact that judges must give reasons for departing upward or downward from the starting point helps to ensure that all relevant personal considerations are canvassed. Judicial discretion remains, but it is less susceptible to exercise on irrelevant or ill-thought-out factors. In short, the starting-point approach does not prevent the judge from considering all relevant personal factors. Rather, it provides a structure which helps to ensure that they are considered and given their proper weight.

(b) Section 7 of the Charter and the Burden of Proof

81Section 7 guarantees that a person may not be deprived of liberty except in accordance with the principles of fundamental justice. One of these principles is said to be the obligation on the Crown to establish a basis for severe punishments. It is argued that the selection of a starting point relieves the Crown of this obligation and forces the accused to prove that he or she is entitled to a lesser punishment.

82The starting-point approach, properly applied, does not relieve the Crown of its burden of establishing the appropriateness of severe punishments. Before the sentencing process begins, the Crown must have established that the accused is guilty of the offence charged. The Crown must then go on, if it has not already done so, to establish circumstances of the offence bringing it within the category of offence meriting a particular starting point. This is another way of saying that the Crown must establish that the offence in all the circumstances falls within a particular range. This fixes the starting point. In order to obtain a harsher sentence, the Crown must establish aggravating circumstances. On the other hand, if mitigating factors are revealed, the sentence will be reduced from the starting point.

(c) The Principle that the Punishment Should Be the Least in the Circumstance

83The answer to this argument is that the starting point, appropriately selected, is the least punishment for the circumstances which usually attend a particular type of offence, on the assumption that the accused is a person of good character with no criminal record. If the accused’s character is even better than might be supposed or if there are other factors supporting a lesser sentence, the accused will receive a lesser sentence. On the other hand, if the Crown proves that the accused’s character is worse than might be supposed, that he has a significant criminal record, or that other exacerbating facts exist, it may be increased. In all cases, the final sentence will be the least that is appropriate in the circumstances.

(d) The Argument that a Starting Point for Major Sexual Assault Constitutes Impermissible Judicial Legislation

84The Criminal Code creates no offence of major sexual assault. It creates a single offence of sexual assault which embraces conduct ranging from a grazing touch or quick kiss to the most serious forms of violent rape. In order to give guidance to sentencing judges, the Alberta Court of Appeal divided the vast global offence of sexual assault into two types, minor and major. Major sexual assault, defined in terms of an affront serious enough to make lasting emotional or psychological injury likely, attracts a starting point of three years. The appellant argues that this amounts to judicial legislation of a new offence -- major sexual assault -- with a penalty of three years imprisonment. This, he asserts, lies beyond the power of the courts. Only Parliament can legislate new offences.

85This argument ignores the fact that crimes defined by the Criminal Code invariably cover a range of acts of varying degrees of seriousness in terms of the harm they are likely to cause to the victim, attracting a range of sentences of varying degrees of severity. Consider, for example, the offence of theft. Theft may be so small as to be trivial -- for example, the taking of a quarter from the landlady’s change purse. Or it may be serious -- for example, stealing millions of dollars from innocent investors. The courts properly fashion different sentences according to the different degrees of seriousness of the theft. They do not thereby create new offences. To meet just this problem, the Criminal Code typically provides a wide range in which the sentence for a particular offence may fall. It has done so in the case of sexual assault. The Criminal Code offence anticipates and provides for the fact that courts will find some violations more serious than others and impose higher or lower sentences accordingly. The starting-point approach, as indicated earlier, is merely a variation on the traditional concept of ranges of sentence for particular types of criminal acts. To recognize a certain type of act as being serious or major, and hence, in the typical case, attracting a sentence in a particular range, is not to create a new crime. It is only to recognize what no one would deny -- that a given category of crime as defined by the Criminal Code is capable of embracing a wide variety of conduct, some more heinous and hence deserving of harsher punishment than others.

86Nevertheless, it is argued, the Court of Appeal possesses no power to lay down a “minimum” sentence of three years for a certain type of conduct. The answer to this objection is that the Court of Appeal has not laid down a minimum sentence. It has set a starting point, nothing more. The sentencing judge may move the sentence down from the starting point if the circumstances support a lesser sentence, just as he or she may move it up. The starting point says no more than this: in a typical case, with an offender of good character and no criminal record, this would be an appropriate sentence. To say this is not to tread on the exclusive legislative domain of Parliament. It is rather to carry out the sentencing mandate that Parliament, in enacting crimes covering a broad variety of conduct accompanied by a broad sentencing range, has imposed on the courts.

(e) The Vagueness Argument

87The appellant argues that over time, the types of conduct included within the term “major sexual assault” has expanded so that what is a major sexual assault is unclear. This means that a person accused of or contemplating sexual assault cannot get clear legal advice on what sort of charge he or she faces. The starting point therefore creates a vague law which fails to delineate the area of risk as required by s. 7 of the Charter.

88The Charter catches vagueness at two points. First, a criminal offence may be so vaguely defined that it contravenes s. 7. Second, a legislative or state limit on a right or freedom may be unjustified because it is overbroad. See Professor D. Stuart, Charter Justice in Canadian Criminal Law (1991), at p. 79. The first concern is at issue here.

89Lamer J. (as he then was) described the basis for the principle as follows in Reference re ss. 193 and 195.1(1)(c) of the Criminal Code, 1990 CanLII 105 (SCC), [1990] 1 S.C.R. 1123, at p. 1152:

The principl[e] ...[is] based on the ancient Latin maxim nullum crimen sine lege, nulla poena sine lege -- that there can be no crime or punishment unless it is in accordance with law that is certain, unambiguous and not retroactive. The rationale underlying this principle is clear. It is essential in a free and democratic society that citizens are able, as far as is possible, to foresee the consequences of their conduct in order that persons be given fair notice of what to avoid, and that the discretion of those entrusted with law enforcement is limited by clear and explicit legislative standards....

90In my view, this principle does not assist the appellant. First, the principle applies to the question of what conduct is criminalized. It neither has nor should be applied to sentencing ranges. The protection against vague criminal legislation entitles a person to know what the prohibited conduct is and the type of sentence it may attract. We have been referred to no authority extending the principle to require certainty within a given sentencing range. To condone such an extension would be unwise, in my view. The Criminal Code contains hundreds of offences with wide penalty ranges. To hold that the law must describe with certainty the precise sentence which particular conduct may attract would be to render all these laws subject to attack on the ground that they are too vague. In place of the existing regime of broadly defined offences with broad ranges of sentences, Parliament would be compelled to legislate thousands of precise crimes attaching precise penalties. While judicial discretion to sentence would be reduced, it may be debated whether the public would in the end be better informed.

91In my view, the principle against vagueness is satisfied if the law specifies the conduct prohibited and indicates a range of punishment. The potential offender may not know precisely what sentence will be imposed for a contemplated criminal act, but he or she knows the risk engaged. If the offender commits the criminal act and accepts that risk, the offender cannot complain. In the context of sexual assault, if the offender chooses to commit a sexual assault which a reasonable person could foresee would be likely to cause lasting emotional or psychological harm, he or she properly bears the risk of a major sentence.

92Secondly, even if the principle against vagueness did apply to sentencing ranges, the starting point here at issue would satisfy it. If anything, the starting-point approach increases the knowledge of the risk faced by a prospective or convicted offender. The offender knows that forcing sexual conduct on another without consent is a criminal act. The offender also knows that this criminal activity attracts a range of punishment and that serious sexual violations attract serious penalties. After Sandercock and the starting-point approach, the offender knows that assaults associated with a reasonable expectation of lasting emotional or psychological harm are considered major and typically attract sentences ranging upward or downward from a starting point of three years, depending on the individual circumstances of the case. This knowledge extends to counsel advising the offender on the conduct of his or her case. Far from being the victim of vagueness, the offender knows more about the risk he faces than ever before.

93The possibility that judges may differ on whether to consider a particular assault major does not attract the principle against vagueness. In Reference re ss. 193 and 195.1(1)(c) of the Criminal Code, supra, Lamer J. made it clear that vagueness will not arise simply because the law is open to varying interpretations by the courts. Flexibility is not synonymous with vagueness. Rather, the question is whether the law has been given sensible meanings by the courts (p. 1157). In that case, the courts were left to define terms found in Criminal Code provisions. The language of the provisions left criminal offences open to a wide variety of meanings. That is not the case here. The criteria for a major sexual assault set forth in Sandercock are clear. The most significant is the requirement that the act be of a type likely to cause lasting emotional or psychological harm. This assessment is well within the capacity of trial judges.

(f) The “Presumption” of Harm

94The appellant argues that once the actus reus for a major sexual assault is proved, there is a presumption that the assault caused lasting emotional or psychological harm to the victim. This, he alleges, would violate the general principle that the Crown must prove all aggravating factors going to sentence. He argues further that a presumption of psychological harm for “major” sexual assaults under s. 271 is inconsistent with the offence of aggravated sexual assault found in s. 272 of the Criminal Code, where the Crown is required to prove bodily harm.

95This argument rests on a misunderstanding of what the starting-point approach involves. The starting-point approach, properly conceived, does not involve a presumption of harm. In determining the starting point, the judge must consider whether the violation of the victim’s integrity was “such that a reasonable person would know beforehand that the victim likely would suffer lasting emotional or psychological injury, whether or not physical injury occurs”: Sandercock, supra, at p. 84 (emphasis added). This involves neither a presumption of harm nor proof of harm that the Crown would be obliged to make under s. 272 of the Criminal Code. If the judge concludes that the act was of a sort which would make lasting emotional or psychological harm likely, the judge may classify the assault as major. But the judge does not presume that the harm has in fact occurred in the particular case on trial. Whether or not harm actually occurred is not an issue at the first step of the process, determining the starting point. Actual harm on the facts of the case becomes an issue only at the second stage of individualization of the sentence. Again, at this stage, there is no presumption of harm. The judge must determine whether in fact the crime caused lasting emotional or psychological harm to the victim. That falls to be determined on the facts. If the Crown wishes to rely on actual harm, it must prove it without the aid of any presumption.

96The Alberta Court of Appeal made this clear in Sandercock. This is also how I understand the comments of the Court of Appeal in the present case. The court stated (at p. 173):

...in each case, there also exists a very real likelihood of psychological harm. Therefore, what must be understood is that it is not necessary that the Crown prove the existence of this kind of harm as a condition precedent to the courts classifying a sexual assault as a major one. Psychological harm is presumed in the absence of evidence to the contrary. That is what this court said in Sandercock. [Emphasis added.]

The court’s statement is unimpeachable until the last two sentences. As the court notes, the Crown need not prove actual psychological harm in order for the judge to conclude that the assault falls within the category of major sexual assault. The inquiry at this stage is general, concerned not with the actual facts of the case but with the sort of harm this type of assault would be likely to cause. Actual harm or its absence is of no relevance at the first stage of determining a starting point. It is in this sense that I understand the court to be saying that at this stage psychological harm is “presumed”. This said, the language of presumption and in particular the additional phrase “in the absence of evidence to the contrary” are confusing and better avoided. In the end, however, the court made it clear that it was reaffirming Sandercock with its two-part test in which actual harm in the case on trial is never presumed.

97I conclude that because it does not presume actual harm to the victim, the starting-point approach does not violate the rule that the Crown bears the burden of proving exacerbating circumstances, nor, contrary to the view taken by Sopinka J., does it conflict with the logic underlying s. 272 of the Criminal Code, which requires proof of actual harm in the particular case.

(g) Problems of Application

98The foregoing considerations establish that the starting-point approach to sentencing theoretically conforms to the Charter and in particular, to the principles of fundamental justice. It is argued, however, that even if the approach is theoretically sound, problems of application lead to unjust results. First, it is suggested, judges sometime apply starting points mechanically without individualizing the sentence. Second, it is feared courts of appeal may fail to defer to the weight trial judges have given to personalized factors in reducing the sentence from the starting point. Finally, difficulties are said to arise in choosing the appropriate starting point for particular conduct.

99The first point is simply addressed. To treat a starting point as the final point without going on to individualize the sentence is to err in law by failing to consider the actual situation of the offender and the particular facts in the case at bar. Trial judges must bear in mind that starting points, like sentencing ranges, are just what the name purports -- starting points. It is essential that they go on to individualize the sentence. If they do not, the sentence may be adjusted upward or downward on appeal.

100The second point -- that courts of appeal may too easily interfere with a trial judge’s application of the starting-point approach -- is also easily addressed. Courts of appeal should not discount individualizing factors which the trial judge properly took into account in the interests of emphasizing uniformity or sending a stronger deterrent message. If they do step, they err in law by ignoring the principle of deference to the sentencing judge. It cannot be stressed too much that the offender may not be sacrificed to the perceived greater good of a standard tariff or to the goal of general deterrence. The offender is entitled to the sentence which his or her crime, conduct and circumstances mandate, no more or less. Courts of appeal also err in law if they interfere with a sentence solely on the ground that the trial judge has not cited the appropriate starting point.

101Courts of appeal may intervene where the sentence discloses error of principle or is demonstrably unfit: R. v. Shropshire, 1995 CanLII 47 (SCC), [1995] 4 S.C.R. 227; R. v. M. (C.A.), 1996 CanLII 230 (SCC), [1996] 1 S.C.R. 500. The starting-point approach does not provide a new judge-made legal principle, enabling courts of appeal to interfere with the proper exercise of the trial judge’s discretion, as Sopinka J. suggests. The starting point merely indicates the appropriate range of sentence for an offence of a certain degree of seriousness. When a court of appeal interferes on the ground that the judge ignored the correct starting point, it is simply saying that the sentence is demonstrably unfit because it falls outside the acceptable range of sentence for that sort of offence. This Court in Shropshire, supra, at paras. 48 and 50, citing R. v. Muise (1994), 1994 NSCA 198, 94 C.C.C. (3d) 119 (N.S.C.A.), expressly acknowledged the need for courts of appeal to intervene where the sentence falls outside the appropriate range:

The most that can be expected of a sentencing judge is to arrive at a sentence that is within an acceptable range. In my opinion, that is the true basis upon which Courts of Appeal review sentences when the only issue is whether the sentence is inadequate or excessive.

Unreasonableness in the sentencing process involves the sentencing order falling outside the ‘acceptable range’ of orders.... [Emphasis added.]

102The third problem -- that of choosing the appropriate starting point for a particular offence -- requires more discussion. It lies at the heart of the difference between the trial judge and the Court of Appeal in this case. The trial judge found that the offences did not fall within the category of “major sexual assault” attracting a starting-point sentence in the range of three years. The Court of Appeal disagreed.

103It is critical that the appropriate starting point be used. To place an offence in a higher category than warranted may result in a sentence which is too harsh. On the other hand, to place an offence in a lower category than warranted may result in a sentence which is disproportionately low having regard to the penalties imposed on other offenders for the same or similar conduct.

104To avoid the problem of inappropriate starting points, courts of appeal must first provide clear descriptions of what type of offence falls within what category. This done, trial judges must properly classify the offence before them having regard to these criteria. This exercise is not new to courts. Judges have long assessed the general seriousness of particular types of offences with a view to determining the appropriate range of sentences.

105But clear descriptions of what suffices to bring a particular offence within a particular category with a particular starting point are not enough. Trial judges and courts of appeal must adhere strictly to its logic. The process, to repeat, consists of two steps. First the starting point must be ascertained. The starting point is based on the general characteristics of the typical offence and on the assumption that the offender is a person of good character with no criminal record. It follows that it is wrong to introduce individualized factors relating to the offender and the victim in assessing the seriousness of the offence in order to determine the starting point. Once the starting point has been ascertained on the basis of the general characteristics of the offence, the inquiry moves to the second stage of individualization. Here the judge considers factors personal to the accused, the victim, and the actual consequences of the offence in order to determine whether the sentence should be higher or lower than the starting point.

106The difficulty in the case at bar stemmed in part from the fact that the trial judge confused these two steps. When determining whether the sexual assault was “major”, she should have looked only to the nature of the offence having regard to the violence or degradation perpetrated, the nature of the act and the likely effect such an act would typically have on its victim. She should not have considered mitigating factors peculiar to the particular offender and victim in the case before her. In particular, the fact that the accused had been drinking and that the victim had pre-existing problems contributing to her trauma were not relevant factors to consider when determining whether the assault was major. Consideration of these factors led her to conclude that the assault was not major and had the effect of imposing a sentence far below the three-year period of incarceration appropriate for a major sexual assault. The result was a sentence which failed to take proper account of the need for a measure of uniformity and equity between sentences for the same or similar offences.

(6) Conclusion on the Appropriateness of the Starting-Point Approach

107I conclude that the starting-point approach, properly understood and applied, is theoretically sound and marks an advance in the need to find a principled approach to the dual goals of individualization of sentences and the need for uniformity and consistency.

B. Application to the Facts of This Case

108The appellant’s contention that the starting-point approach has no place in sentencing must be dismissed. The remaining question is whether the Court of Appeal properly applied that approach in the case at bar.

109The appellant argues that the choice of a starting point is a matter solely within the discretion of the trial judge and that the Court of Appeal should not have interfered with her conclusion that the offences were not major sexual assaults. This argument, with respect, is misconceived. As indicated earlier, the “starting point” is not a principle of law, but rather a tool to determine the proper range of sentence for a certain type of offence. Failure to allude to the appropriate starting point or range is not an error of principle as that term is used in M. (C.A.), supra. If the trial judge fails to refer to the appropriate starting point or range but in the end imposes a sentence within the acceptable range of sentence for the offence as adjusted for the particular circumstances of the offender, a court of appeal should not interfere. On the other hand, if the sentence falls outside the appropriate range, the court of appeal must interfere: Shropshire, supra.

110In my view, the trial judge’s sentence was “demonstrably unfit”, and “in substantial and marked departure from the sentences customarily imposed for similar offenders committing similar crimes”(M. (C.A.), supra, at paras. 90 and 92). The assaults at issue in the case at bar shared the following characteristics: (a) they involved young victims; (b) the accused held a position of trust in relation to the victims; and (c) the first assault involved fondling and attempted rape and the second involved fondling. A review of many cases from Alberta reveals that a sentence of less than two years in circumstances similar to the case at bar constitutes a marked departure from the norm. See: Sandercock, supra; R. v. S.G.O.R. (1991), 113 A.R. 36 (C.A.); R. v. Sand (1991), 1991 CanLII 13091 (AB CA), 120 A.R. 397 (C.A.); R. v. S. (W.B.); R. v. P. (M.) (1992), 1992 CanLII 2761 (AB CA), 73 C.C.C. (3d) 530 (C.A.); R. v. R.S.B. (1992), 1992 ABCA 258, 135 A.R. 23 (C.A.); R. v. Spence; R. v. F. (D.L.) (1992), 1992 ABCA 352, 78 C.C.C. (3d) 451 (C.A.); R. v. R.E.C., [1993] A.J. No. 303 (C.A.); R. v. D.M.B. (1993), 1993 CanLII 16356 (AB CA), 141 A.R. 307 (C.A.); R. v. Nicholson (1993), 1993 ABCA 248, 145 A.R. 262 (C.A.); R. v. Dionne, [1993] A.J. No. 939 (C.A.); R. v. D.K. (1994), 1994 ABCA 219, 155 A.R. 269 (C.A.); R. v. Watson (1994), 1994 ABCA 234, 157 A.R. 80 (C.A.); R. v. Wagar (1995), 1995 ABCA 413, 174 A.R. 317 (C.A.); R. v. Lapatak (1995), 1995 ABCA 322, 169 A.R. 385 (C.A.); and R. v. Lakotos (1996), 1996 ABCA 262, 187 A.R. 45 (C.A.). The case of R. v. A.B.C. (1991), 1991 ABCA 337, 120 A.R. 106 (C.A.), where a sentence was imposed of one year in jail and one-year probation, while superficially similar, is distinguishable on a number of grounds, including the fact that the offender and the victim sought counselling together after the assault, expert evidence suggested no risk of re-offending, and the occurrence was a “one-time” isolated event.

111The unfit nature of the sentence resulted from the trial judge’s failure to place appropriate weight on the appropriate factors, particularly the impact this sort of offence may have on the victim. This in turn, depreciated the seriousness of the offence and led to a demonstrably unfit sentence. The trial judge held that there was no violence or threats, but failed to consider whether sex obtained from a sleeping minor to whom one stood in a relationship of trust was equivalent to threats. Most significantly, she did not ask herself whether this was the sort of offence which would likely result in lasting psychological or emotional harm, the hallmark of a major sexual assault. She placed great emphasis on the physical acts, for example that the penetration on the first assault was not complete, but none on the violation of the victim’s integrity and sense of self-worth and control over her body and the likely effect that this sort of criminal act would have on her emotional state. This amounted to an overemphasis on some factors and underemphasis on other factors. It also led to a demonstrably unfit sentence. The Court of Appeal was therefore correct to intervene: M. (C.A.), supra; Shropshire, supra.

112I conclude that the Court of Appeal had grounds to set aside the sentence imposed by the trial judge. The remaining question is whether the sentence it substituted of four years’ imprisonment for the first offence followed by one year for the second offence was too high. In my view, it was not.

113The Court of Appeal correctly identified the first offence as a major sexual assault which normally in the case of an offender of good character and no record would attract a sentence in the range of three years. While it involved no violence or threats, it involved conduct equally abusive and calculated to cause lasting harm -- the breach of trust of a young girl in the appellant’s care. The assaults occurred when the young girl was helplessly sleeping, in circumstances where she could not have feared violation. The result of such an assault on a typical victim would likely have been shame, embarrassment, unresolved anger, a reduced ability to trust others and fear that even in innocent sleep, people could and would abuse her and her body. In short, lasting emotional and psychological harm was likely. Viewing the assault objectively in terms of its probable harmful consequences, it was serious and major.

114The Court of Appeal went on to move the sentence for the first assault up from the three-year starting point to four years. At this stage of the analysis the court must look at particular factors which differentiate the case at bar from the typical case contemplated by the starting point. The case on appeal exhibited few mitigating factors and a number of aggravating factors. The appellant’s good character and lack of criminal record is not a mitigating factor since this is assumed in the typical case posited by the starting point. The fact that he was drunk when he committed the offence and did so “spontaneously” is arguably a mitigating factor, but not one to which I would attach great weight. Sexual assaults are often unplanned and spontaneous, and drunkenness to the point that one cannot control one’s sexual urges is itself morally blameworthy. The evidence confirmed that the victim had suffered actual harm, ruling out mitigation of sentence on the basis that the victim had not in fact been traumatized. In favour of mitigation might be cited the fact that the appellant and his family would likely suffer if he were incarcerated for a prolonged period. On the other side of the ledger lie the appellant’s callous abuse of trust and his unconcern with the effect his acts would likely have on his victims. The victim was very young and entirely vulnerable. The fact that she had already been damaged should have increased the concern and respect the appellant held for her. The evidence showed that she had in fact suffered lasting emotional and psychological harm as a result of the assault. It was open to the Court of Appeal on these facts to conclude that there was little basis to move down from the typical sentence of three years and considerable basis to impose a higher sentence. In the end, the Court of Appeal imposed a sentence which reflected the seriousness of the offence as well as the particular circumstances of the offender.

115The second offence, while arguably less serious in that there was no actual sexual intercourse, also involved violation of a young girl in circumstances of trust. She had been invited into the appellant’s home as a babysitter, and was entitled to suppose that he would protect her rather than violate her while she was there. Instead he attacked her in an intimate and sexual fashion as she slept, forcing her to flee his house in the night. The likely consequence of such behaviour is that it will impair a victim’s ability to form trusting relationships and live a life free of fear of violation, and leave her feeling defiled, degraded, and full of unresolved anger. This suffices to qualify the assault as major and bring it within the range of a three-year sentence, were it viewed in isolation.

116This brings me to the question of the relationship between the two offences and the effect that this should have on the global sentence. The trial judge ordered that they run concurrently. In my view, the Court of Appeal was correct to conclude that concurrent sentences were not appropriate in this case. The case involved two separate offences on two different victims, separated by a period of seven years. Having determined that the sentence imposed by the trial judge was demonstrably unfit, it was open to the Court of Appeal to reconsider all aspects of the sentence, including the relationship between the two offences involved. Nevertheless, the global effect of the sentences must be considered. In my view, the Court of Appeal correctly imposed a sentence of one year for the second offence, to be served consecutively.

117I would dismiss the appeal and confirm the sentence.