Court of Criminal Appeal Supreme Court New South Wales Medium Neutral Citation: PG v R [2017] NSWCCA 179 Hearing dates: 8 March 2017 Decision date: 28 July 2017 Before: Basten JA at [1]; Button J and N Adams J at [70] Decision: 1. Grant the applicant leave to appeal from the judgment on sentence of the District Court delivered on 17 November 2014. 2. Dismiss the appeal. Catchwords: CRIME – sentencing – parity principle – applicant's undiscounted sentence significantly greater than co‑offender's – whether justifiable sense of grievance – whether appropriate to compare undiscounted sentences – whether evidential differences justified disparity in initial sentences CRIME – sentencing – application of discounts for guilty pleas under Crimes (Sentencing Procedure) Act 1999 (NSW) s 22 when imposing aggregate sentence under s 53A – whether appropriate to apply discount to aggregate sentence or putative individual sentences Legislation Cited: Crimes (Sentencing Procedure) Act 1999 (NSW), ss 3, 4, 21A, 22, 24, 53A; Pt 2, Div 1; Pt 3 Criminal Appeal Act 1912 (NSW), s 6 Interpretation Act 1987 (NSW), ss 8, 33 Crimes (Sentencing Procedure) Amendment Act 2010 (NSW), Sch 2 [14] Cases Cited: Bao v R [2016] NSWCCA 16 Elsaj v R [2017] NSWCCA 124 Glare v R [2015] NSWCCA 194 Green v The Queen (2011) 244 CLR 462; [2011] HCA 49 Hampton v R [2014] NSWCCA 131 JL v R [2014] NSWCCA 130 JM v R [2014] NSWCCA 297 Khawaja v R [2014] NSWCCA 80 Lowe v The Queen (1984) 154 CLR 606 Martin v R [2014] NSWCCA 124 Postiglione v The Queen (1997) 189 CLR 295 R v Cahill [2015] NSWCCA 53 R v Nykolyn [2012] NSWCCA 219 R v SY and Anor [2003] NSWCCA 291 R v Thomson and Houlton (2000) 49 NSWLR 383; [2000] NSWCCA 309 R v Borkowski [2009] NSWCCA 102; 195 A Crim R 1 SHR v R [2014] NSWCCA 94 Stoeski v Regina [2014] NSWCCA 161 Subramaniam v R [2013] NSWCCA 159 Category: Principal judgment Parties: PG (Applicant) Regina (Respondent) Representation: Counsel: M Avenell (Applicant) E Balodis (Respondent) Solicitors: Legal Aid NSW (Applicant) Solicitor for Public Prosecutions (Respondent) File Number(s): 2014/30181; 2014/30184 Decision under appeal Court or tribunal: District Court Jurisdiction: Criminal Date of Decision: 17 November 2014 Before: Wells SC DCJ File Number(s): 2014/30181; 2014/30184 Judgment 1. BASTEN JA: On 17 November 2014 the applicant was sentenced by Judge Wells SC in the District Court at Lismore on an indictment containing nine serious criminal charges, with a further seven offences to be taken into account with respect to count 1. The offending took place on two separate dates, namely 6 October 2013 and 16 November 2013. The applicant was sentenced to an aggregate term of 9 years imprisonment with a non-parole period of 6 years 9 months, commencing on the date of his arrest, namely 30 January 2014. 2. The sentencing was undertaken on the basis of pleas of guilty and a statement of agreed facts. The applicant gave brief evidence on the sentencing hearing, which was not entirely consistent with the agreed facts. In setting out the indicative sentences for the various offences, the sentencing judge gave effect to a discount of 25% against the sentence that might otherwise have been imposed on account of the early plea of guilty. She allowed a further 25% for assistance to authorities, giving a total discount of 50%. Issue on appeal – parity 1. There was no appeal against that sentence, nor on the present application for leave to appeal is any issue taken with the propriety of the sentence, looked at in isolation. Rather, the present application is based on an alleged disparity between the sentence imposed on the applicant and that later imposed by the same judge on the principal co-offender, Brandon Green. The aggregate sentence imposed on Mr Green on 4 April 2016 also involved imprisonment for 9 years, with a slightly shorter non-parole period, namely 6 years 4 months. Mr Green had the benefit of a 25% discount for an early plea of guilty, but no discount for assistance. 2. On 14 November 2016 the solicitor for the applicant filed a notice seeking leave to appeal containing one ground, namely that "[t]he applicant has a justifiable sense of grievance when comparing the sentence imposed upon him with that imposed upon his co-offender Mr Green." 3. The basis of the appeal may be shortly stated. Mr Green's undiscounted aggregate sentence would have been imprisonment for 12 years with a non-parole period of 8 years 5 months. The undiscounted aggregate sentence which would have been imposed on the applicant was 18 years with a non-parole period of 13 years 6 months. So viewed, the sentencing judge has treated the culpability of the applicant as 50% higher than that of Mr Green, or, from the other perspective, Mr Green's culpability as two-thirds that of the applicant. While the applicant accepted that there were reasons justifying a higher sentence in his case, he submitted that the differential was grossly disproportionate to the levels of comparative culpability. 4. The difficulties faced by the applicant in this regard were significant. First, both sentencing exercises were undertaken by the same judge. Secondly, the question of parity was alive in the minds of the prosecutor and the three co-offenders who were sentenced on 4 April 2016; the judge expressly compared the applicant and Mr Green on numerous occasions in the course of sentencing Mr Green. 5. Thirdly, it is not, on any view, appropriate to compare the two undiscounted sentences. Both received a 25% discount for their pleas of guilty, but the applicant received a further 25% discount for assistance. Taking the latter discount into account, but not the common discount, the applicant's starting point was 13.5 years, compared with 12 years for Mr Green. On that approach, the relevant comparison showed a starting point for the applicant which was 12.5% higher than that of Mr Green. 6. Accordingly, in the absence of some significant identifiable error, the only basis on which the applicant seeks leave to appeal is in order to review the evaluative judgment of the relative culpabilities of each offender. Sentencing principles – parity 1. It is convenient to identify aspects of the parity principle relevant to the present application. An important element in determining the present application is that the applicant was sentenced in a separate proceeding from his co-offenders. Certain significant aspects of the evidence before the sentencing judge were not identical in each proceeding. 2. The "parity principle" has proved contentious, both as to its scope and the consequences of non-compliance. That is because it provides a basis for intervention on an offender's appeal against severity, in circumstances where the sentence imposed (as in this case) is otherwise appropriate in the sense that it does not call for appellate intervention. (Different issues will arise in prosecution appeals, which need not be addressed here.) In New South Wales, the principle is addressed under s 21A of the Crimes (Sentencing Procedure) Act 1999 (NSW) ("Sentencing Procedure Act"), which specifies matters which the court is required to take into account, "in addition to any other matters that are required or permitted to be taken into account by the court under any Act or rule of law." [1] 3. It therefore matters, as Bell J pointed out in Green, how widely the concept of parity as between co-offenders is to be understood. [2] That precise point does not arise in the present case, as each of the applicant and Mr Green were charged with the same offences, arising out of the same course of criminal activity. Rather, the present appeal turns on the factors with which the parity principle engages. Again adopting the language of Bell J in Green: [3] "The principle of parity in sentencing, stated by this Court in Lowe v The Queen, applies to the sentencing of co-offenders whose culpability for the offence and whose antecedents are comparable. [4] Where there is a marked disparity in the sentences imposed on co-offenders engendering a justifiable sense of grievance, the appellate court is entitled to intervene and to reduce an otherwise appropriate sentence. [5] " 1. Both of these elements require explanation. In Lowe v The Queen, Mason J spoke of intervention in a case where the "disparity engenders a justifiable sense of grievance in the applicant and an appearance of injustice to that impassive representative of the community, the objective bystander." [6] Such language explains the justification for intervention, but not the standard to be applied. As explained in the joint reasons in Green: [7] "The sense of grievance necessary to attract appellate intervention with respect to disparate sentences is to be assessed by objective criteria. The application of the parity principle does not involve a judgment about the feelings of the person complaining of disparity. [8] " 1. As to the standard to be applied, Gibbs CJ stated in Lowe: [9] "It is obviously desirable that persons who have been parties to the commission of the same offence should, if other things are equal, receive the same sentence, but other things are not always equal, and such matters as the age, background, previous criminal history and general character of the offender, and the part which he or she played in the commission of the offence, have to be taken into account." 1. In Lowe, Brennan J said that "[t]o facilitate the comparison of conduct and antecedents it is desirable that, where practicable, co-offenders be sentenced by the same judge at the same time." [10] That did not happen in Lowe, Lowe and his co-offender being sentenced by different judges on different occasions. The judge sentencing the co-offender found that the co-offender's position was "sufficiently different" from that of Mr Lowe to allow the imposition of a lesser sentence. Referring to the judgment in the Queensland Court of Appeal, Brennan J stated: [11] "Campbell CJ, in whose judgment in this respect Macrossan and Derrington JJ agreed, found that there were 'differences of considerable weight and significance between the two cases, particularly when one looks at the parts played and the material placed before each of the sentencing judges in relation to each of the co-accused'. The difference identified in the judgment of Campbell CJ is that 'Lowe was the prime mover in the commission of this offence'." 1. It would appear from this passage that the sentencing judges did not each have the same material before them. However, Brennan J continued: "A reading of the application book does not satisfy me that their Honours were right to find that Lowe's conduct was so much graver than Smith's that both sentences can be supported. Before Thomas J, Smith's counsel had argued that Lowe was the main actor but it had not been part of the case alleged against or admitted by Lowe. Indeed Kelly J, in passing sentence on Lowe, had said 'it does seem that you and Smith were at least equally involved in its preparation and some thought was clearly given to the carrying of it out'. Although the Court of Criminal Appeal was entitled to give appropriate weight to the argument advanced in Smith's case that Lowe was the main actor, it seems to me that there is such a disparity between the sentences as to show that, if Smith's sentence was appropriate, the sentence imposed on Lowe was manifestly excessive." 1. The treatment of the different material before each sentencing judge was dealt with more briefly, or not at all, by other members of the Court. Mason J, who with Brennan J was in a minority in considering that the Court of Criminal Appeal had erred, stated: [12] "In the applicant's appeal, the Court of Criminal Appeal appears to have considered that he was 'the prime mover' in the commission of the offence. However, this was not a fact established by proof or admission before the trial judge. All that the trial judge said in sentencing the applicant was that he and Smith were at least equally involved in the preparation of the crime and in the thought which was given to its execution. Although the Court of Criminal Appeal may have been entitled to proceed on the footing that the applicant's participation was in some degree greater than Smith's, they were not entitled to conclude that there was a marked difference between them." 1. Postiglione [13] also involved two co-offenders sentenced at different times by different judges. That situation had arisen because Mr Postiglione had entered a plea of guilty, whereas his co-offender, Savvas, had stood trial and was sentenced following conviction by Grove J. In sentencing Mr Postiglione, Mathews J concluded that his co-offender was the prime mover and was "calling the shots". [14] Perhaps curiously, McHugh J stated: [15] "On the material before this Court, it is impossible to say whether Grove J or Mathews J reached the correct conclusion concerning Savvas's guilt." Assuming, however, that that did not matter, McHugh J continued: "Thus, the difference in the sentences between the appellant and Savvas is explicable by reference to identifiable factors in the sentencing process which objectively support the different approaches in the two cases." 1. Gummow J also noted that different views had been taken as to the relative culpability of each offender. [16] Gummow J then stated: "The question for this Court is whether there was any error made by the Court of Criminal Appeal in its consideration of the parity principle as applied to the sentences of the appellant and Savvas on the basis of the findings made by Mathews J as to, among other things, their respective degrees of criminality." 1. There were other factors, however, which warranted the conclusion that no sufficient disparity was demonstrated in the circumstances. The question of relevant culpabilities was not determinative. 2. McHugh J and Gummow J were in a minority with respect to the application of the parity principle. The joint judgment of Dawson and Gaudron JJ referred to the disparate evaluations, noting that Grove J treated their levels of culpability as "generally equivalent" although he harboured a suspicion that Savvas was the principal. [17] The joint reasons continued: [18] "That suspicion was in line with the finding by Mathews J when sentencing Postiglione that Savvas was the principal organiser of both conspiracies. The latter finding was accepted by the Court of Criminal Appeal and it is that finding upon which this Court must act." 1. The other member of the majority, Kirby J, appears to have assumed that the difference arose from the manner in which the principle of totality was applied (not parity) [19] and also treated Mr Savvas as having been at the "head of the pyramid", presumably bearing the primary responsibility for the offending. [20] 2. Although the reasoning in each of Lowe and Postiglione dealt with differing assessments of the culpability of co-offenders in a single criminal activity, neither is entirely satisfactory in its consideration of the different factual bases upon which each sentence was imposed. For example, the fact that Mr Savvas was sentenced after Mr Postiglione, and on the basis of generally equal culpability, should have meant that, other things being equal, he would have received a similar sentence to Mr Postiglione, although Postiglione himself was sentenced on the basis that Savvas should have had a higher sentence. In other words, the different factual findings would have explained the relative seriousness of each sentence. That was not, in terms, how the matters were disposed of, although the minority judges in Postiglione (McHugh J and Gummow J) were closest to that reasoning. 3. As will be explained shortly, the issue is of significance in the present case, because, although each co-offender was sentenced by the same judge, sentencing occurred on the basis of different statements of agreed facts and different evidence. It is by no means uncommon that, where pleas are negotiated and the basis of the plea is agreed, A will be sentenced on the basis that B was the principal and A played a lesser role, whereas B will be sentenced on the basis that A was the principal and it was B who played the lesser role. Usually, there will be no mechanism by which a sentencing judge can resolve such contradictory propositions. Nor is it open to this Court to do so. Accordingly, two co-offenders may be sentenced on entirely different bases. It is not open to A (having been sentenced first) to say, "I was sentenced on the basis that B was the principal, but B got the same sentence that I did and therefore my sentence must be reduced." 4. Sentencing does not take place on the basis of objective singular established truth, but on the basis of the evidence before the sentencing judge. Where, as is desirable, the one judge sentences all co-offenders, the urge for equal justice may seem to be frustrated by such factual inconsistencies, that does not give rise to a basis for intervention by this Court. Consideration of bases of sentencing 1. The applicant's submissions focused squarely on the arithmetical disparity between the two sentences. It was recognised, however, that a level of difference was warranted, but not the extent adopted by the trial judge. 2. The following factors found by the sentencing judge warranted the applicant receiving a longer sentence, namely that the applicant: 1. had a more serious criminal record; 2. was more than four years older than Mr Green; 3. had significantly inferior prospects for rehabilitation, and 4. had a higher level of objective criminality than did Mr Green with respect to the second round of offending in November. 1. Although something more should be said by way of justification in relation to these factors, no purpose is served in dealing paragraph by paragraph with the findings of the sentencing judge. It is necessary, however, to identify the significant differences between the agreed facts in each matter. (a) criminal record and rehabilitation 1. Prospects of rehabilitation usually depend in part upon the sentencing judge's assessment of past attempts at rehabilitation, where there has been significant prior offending. The judge described the applicant's criminal record as "very serious for his age", noting that he had been in prison for lengthy periods for similar offences of breaking, entering and stealing, and that the only recent change in his criminal history was that "the level of violence … is escalating in seriousness." [21] With respect to his use of illicit drugs, the judge noted that he had "completed drug rehabilitation programs while in custody" but "has not shown any particular inclination to engage in rehabilitation … when he is not in custody." [22] 2. The judge noted his expressions of remorse, but rejected his attempt to justify his offending on the basis that he had been sexually abused as a boy and had sought revenge on the primary victim of the brutal assaults, who a co-offender had alleged was a paedophile. The judge further noted that the November offences diminished "to some considerable extent his claims of remorse for involvement in [the October offences]." [23] In her concluding remarks, the judge stated: [24] "I have taken into account his subjective factors which involve a difficult upbringing but also a poor record of serious prior offences and little in terms of real rehabilitation." 1. In sentencing Green and two other co-offenders, David John Reynolds and Andrew John Wren, the judge stated: "It is clear from the material that is before the Court at this stage that [the applicant] is properly regarded as the ringleader in the commission of the offences. He had quite a dreadful criminal record including offences of this kind. He had served lengthy terms of imprisonment in relation to such offences. … Green had a limited criminal record, with the most serious being an offence of affray in 2011 for which he served a short period of imprisonment. There are some other less serious offences on his record. It is relevant that he was aged 21 at the time of these offences, compared to that of [the applicant] who was aged 26." 1. The judge then stated with respect to Green: [25] "He expressed regret and remorse for his participation in these offences both to the psychologist and in his sworn evidence to the Court. The psychologist was of the view though that he had limited insight into his behaviour and unfortunately continued to portray himself as the passive figure in the planning of the offences. … His prospects of rehabilitation have some positive aspects in terms of his obvious capacity for further education and contributing in a meaningful way to the community. However, there are also a number of aspects of his behaviour that cause concern. Although I accept that he is remorseful for his participation in these offences, unlike [the applicant] his remorse came at rather a late stage." 1. In sentencing the co-offenders, the judge stated: [26] "I have already noted that the offender [the applicant], who has been sentenced by this Court, has a more significant criminal record and objectively speaking, took a far more serious and proactive role in the commission of these offences. Of course in his favour, his remorse for his involvement must be regarded as more significant, given the fact that he made earlier and more detailed admissions, though I note that they were made at a stage after initial denials and when it became apparent to him that the police were building a reasonable case against him. … In summary then in terms of parity between these offenders, [the applicant] I regard, as I have said, as the prime mover. He had prior experience and involvements in similar serious offences and certainly inflicted most of the damage and pain during the course of the second incident in November 2013. Objectively speaking, he and the offender Green are on similar footing in relation to the offences in October 2013. Aside though from [the applicant's] early plea and assistance, Green has a stronger subjective case in that he was aged only 21 at the time of these offences and had a less significant criminal record. The Court, as a result, would give much greater weight to considerations of rehabilitation in relation to Green, when considering the principles of sentencing that are applicable." 1. On their face, these findings warrant the imposition of a significantly more severe sentence on the applicant, than that imposed on Mr Green. (b) statements of agreed facts 1. For the most part, the statements of agreed facts were similar in each matter, but in two respects they differed. First, the facts agreed to by the applicant included the following description of the attack on the victim in the course of the October house invasion at his premises: [27] "At one stage [the applicant] says he saw that Green had the victim's pants down and appeared to be trying to put a sword (with a sheath on it) into the victim's 'arse' (sic). Whilst this was happening the victim was kicking back saying that no matter what he had done he'd never treated anybody as bad as this. Green replied 'I don't give a fuck.'" 1. In the course of giving evidence on his own sentencing hearing, Mr Green denied that he had at any stage used a sword to assault the victim, by pushing it into his anus. [28] He also denied seeing any such conduct. 2. Secondly, the agreed facts in the applicant's matter contained the following statements: [29] "After [the applicant] had finished loading the gear he went back up the stairs and told Green they were ready to go. Green then obtained a can of petrol from the victim's garage and commenced to pour the petrol on the floor of part of the upstairs section of the house. Green continued to pour the petrol down the stairs to the garage and around the van. [The applicant] asked Green what he was doing. Green said 'I'm setting it on fire.' [The applicant] asked him why. Green said 'to get rid of the evidence.' [The applicant] then opened the roller door and pushed the van out of the garage. Green then set fire to the petrol. One of the offenders closed the door into the garage. They then drove off. Green was the driver." 1. The account of this incident as it appeared in Mr Green's statement of agreed facts was as follows: [30] "After they had finished loading the stolen property the one of the offenders [sic] used petrol from a jerry can located in the premises to set fire to the brick section of the house. The offenders then opened the roller door and pushed the van out of the garage. One of the offenders closed the garage door. They then drove off. Green was the driver." 1. In his evidence, Mr Green reversed those allegations. The following exchange took place in his evidence-in-chief: [31] "Q. Did you set the property on fire? A. No. Q. Did you see the property set on fire? A. Yes. Q. Where were you when you saw the property set on fire? A. In the driver's seat of [the victim's] van. Q. Where was the van at that stage? A. That was in his driveway. Q. What happened once you saw that property on fire, what happened next? A. [The applicant] jumped in the passenger side of the van, closed the door and we drove off." 1. He was further asked if they had any plan to set the house on fire, which he denied. [32] 2. In cross-examination, Mr Green agreed that he first gave an interview to the police on 7 January 2015, after the applicant had been sentenced. He said it was dangerous to give an account to the police. The following exchange then took place: [33] "Q. So why did you talk to the police then? A. Because I didn't – I didn't belief that everything [the applicant] was saying was true. Q. Well look the situation is you were aware that [the applicant] had been sentenced. You are aware of what he said? A. Yes. Q. He blamed you for lighting the fire and you blame him now for lighting the fire? A. Yes. Q. Isn't it the case really that you're effectively paying [the applicant] back? A. No." 1. In sentencing Mr Green, the judge was not able to be satisfied as to which person was responsible for splashing petrol around or setting fire to the house. [34] Accordingly, the case was left as one of joint criminal enterprise, in which Green did nothing to prevent the fire from occurring or seek to help or assist the victim. The judge continued: [35] "Green, in his evidence to this Court, was simply unable to tell the Court why he did nothing when he was aware that a fire had been lit. His evidence, it should be noted, was that by the time he was aware that the fire was lit, he was seated in the van downstairs ready to leave and his focus was on getting away from the premises." 1. There was no finding as to the allegation which appeared in the applicant's proceedings that Mr Green had indecently assaulted the victim by seeking to thrust a sheathed sword into his anus. It should be accepted that this conduct was not relied upon in sentencing Green. 2. The significance of the differences in the accounts is that, had Mr Green been sentenced on the basis of the agreed facts as were accepted in the applicant's proceedings, he would no doubt have got a significantly longer sentence. As a result of his denials, largely unchallenged by the prosecutor, he was sentenced on a basis which diminished his culpability, in two respects. (c) challenges to factual findings 1. The applicant sought to diminish the disparity in levels of culpability, as found by the sentencing judge, in relation to two matters. First, counsel submitted that, although the applicant was said, in some passages, to be 26 years of age at the time of the offending, he did not in fact turn 26 until some two weeks after the second offences. On the other hand, it was said that although Mr Green was 21 years old, in fact he was nearly 22 years, in that his next birthday was in February 2014. These differences are trivial and could not have affected the assessment of the respective levels of culpability. 2. Secondly, it was said that the judge, in sentencing Mr Green, had made a factual mistake in concluding that he had fallen under the influence of the applicant and had told a psychologist "that he participated in the initial offence after being nagged by his co-offender [the applicant]." [36] The applicant submitted that the person who had nagged Green was not the applicant but the co-offender, Reynolds. It was Reynolds to whom Mr Green had referred in the passage recorded in the psychologist's report. 3. Accepting that it was Mr Reynolds and not the applicant who had "nagged" Mr Green, there was other evidence which supported the conclusion that the applicant was the dominant character and played a leading role in the criminal activity in both October and November 2013. There is no reason to suppose that the reference to the applicant, rather than Reynolds, nagging Mr Green affected Green's sentence. In any event, that fact cannot have affected the sentencing of the applicant because it was not part of the evidence at his sentencing. Conclusions 1. It is not necessary to address the detail in which the sentencing judge assessed the objective seriousness of the various separate offences with respect to the applicant and Mr Green. It is sufficient to note that, in respect of the November offending, she ascribed a significantly higher level of culpability to the applicant. That, taken with his longer criminal record and the judge's assessment of prospects of rehabilitation, warranted the disparity in the sentences. 2. Further, although the disparity would have been reduced had Mr Green been sentenced on the agreed facts that were before the Court in relation to the applicant, each sentence must be considered, as the sentencing judge was required to consider it, by reference to the evidence before the court in sentencing the particular offender. Differences in the evidence justify, and may demand, differences in the sentences imposed. 3. Because the questions relating to the proper application of the parity principle are of more general importance, there should be a grant of leave to appeal. Nevertheless, on analysis of the respective sentencing proceedings, the complaint of unwarranted disparity is without merit and the appeal should be dismissed. Discounting an aggregate sentence 1. My colleagues' response to the above discussion, in which I have assumed that it is appropriate to discount an aggregate sentence, requires that some further explanation be provided as to why, in my view, that is the case. (a) statutory construction 1. The question is to be addressed first as a matter of statutory construction. An aggregate sentence has been an available option since 14 March 2011 pursuant to s 53A of the Sentencing Procedure Act. [37] 53A Aggregate sentences of imprisonment (1) A court may, in sentencing an offender for more than one offence, impose an aggregate sentence of imprisonment with respect to all or any 2 or more of those offences instead of imposing a separate sentence of imprisonment for each. (2) A court that imposes an aggregate sentence of imprisonment under this section on an offender must indicate to the offender, and make a written record of, the following: (a) the fact that an aggregate sentence is being imposed, (b) the sentence that would have been imposed for each offence (after taking into account such matters as are relevant under Part 3 or any other provision of this Act) had separate sentences been imposed instead of an aggregate sentence. (3) Subsection (2) does not limit any requirement that a court has, apart from that subsection, to record the reasons for its decisions. (4) The term, and any non-parole period, set under this Division in relation to an aggregate sentence of imprisonment is not revoked or varied by a later sentence of imprisonment that the same or some other court subsequently imposes in relation to another offence. (5) An aggregate sentence of imprisonment is not invalidated by a failure to comply with this section. 1. The term "sentence" is defined in s 3 in the following terms: 3 Interpretation (1) In this Act: … sentence means: (a) when used as a noun, the penalty imposed for an offence, and (b) when used as a verb, to impose a penalty for an offence. 1. Because the term "sentence" refers to the imposition of a "penalty", it may be accepted that the imposition of a penalty in turn will be a "sentence". Part 2 of the Act is headed "Penalties that may be imposed"; Div 1 contains the following provision: 4 Penalties generally (1) The penalty to be imposed for an offence is to be the penalty provided by or under this or any other Act or law. (2) The penalty to be imposed for a statutory offence for which no penalty is so provided is imprisonment for 5 years. (3) Part 3 applies to the imposition of all penalties imposed by a court, whether under this Act or otherwise. 1. Part 3 of the Sentencing Procedure Act, referred to in s 4(3), includes ss 21-43. The operative terms of Pt 3 involve differing terminology, which may, for present purposes, fall into two categories. One category involves a focus on "an offence", such as s 21A, which applies when "determining the appropriate sentence for an offence" and identifies the relevant aggravating and mitigating factors. A second category focuses on "the offender", such as s 24 which identifies certain matters which are to be taken into account "[i]n sentencing an offender". 2. The present issue arises with respect to the operation of s 22, which appears in Pt 3, and falls within the first category: 22 Guilty plea to be taken into account (1) In passing sentence for an offence on an offender who has pleaded guilty to the offence, a court must take into account: (a) the fact that the offender has pleaded guilty, and (b) when the offender pleaded guilty or indicated an intention to plead guilty, and (c) the circumstances in which the offender indicated an intention to plead guilty, and may accordingly impose a lesser penalty than it would otherwise have imposed. (1A) A lesser penalty imposed under this section must not be unreasonably disproportionate to the nature and circumstances of the offence. (2) When passing sentence on such an offender, a court that does not impose a lesser penalty under this section must indicate to the offender, and make a record of, its reasons for not doing so. (3) Subsection (2) does not limit any other requirement that a court has, apart from that subsection, to record the reasons for its decisions. (4) The failure of a court to comply with this section does not invalidate any sentence imposed by the court. 1. The question is whether s 22 must be applied in imposing an aggregate sentence. There are three reasons for thinking that it may not apply. First, it operates with respect to sentencing for "an offence", whereas s 53A only applies to sentencing "for more than one offence". Secondly, s 22 only applies where the offender has pleaded guilty; s 53A may be invoked in circumstances where the offender has pleaded guilty to some offences, but has gone to trial on others. It is not clear how s 22 should operate in that situation. Thirdly, s 53A(2) requires that the court, in imposing an aggregate sentence, should indicate the individual sentences which would otherwise have been imposed, had the offender been separately sentenced for each offence. In determining the putative individual sentences, s 53A(2)(b) requires that the judge take into account such matters as are relevant under Pt 3, an exercise that will include applying s 22, so that there could be double- counting if the same exercise were to be undertaken with respect to the aggregate sentence. 2. Acknowledging each of these considerations, there are powerful countervailing considerations. It is convenient to start with the terms of s 53A. First, it involves the imposition of a penalty by a court and therefore is an exercise to which Pt 3 applies, pursuant to s 4(3); no other authority is required to engage Pt 3. Section 53A does not expressly disengage s 22, or other provisions in Pt 3. 3. Secondly, s 53A does not disengage s 22, or other provisions in Pt 3, by implication. The requirement to record putative individual sentences would not of itself attract the operation of Pt 3, because these putative individual sentences are not sentences imposed on the offender. Therefore, in order to ensure that they are properly assessed in accordance with Pt 3, it is necessary to make specific provision to that effect, as does s 53A(2)(b). Thus, there is no exclusionary inference to be drawn from such a reference with respect to putative individual sentences which would deny the application of Pt 3 in relation to the aggregate sentence. 4. Thirdly, the fear of double-counting is misconceived. If, in indicating the putative individual sentences, s 22 is applied in its terms, the court will, in a conventional manner, indicate the discount for the guilty plea in each case, thus also indicating the undiscounted starting point. In fixing the aggregate sentence, it is therefore easy to take into account both figures appropriately. 5. There are, in any event, other circumstances in which a risk of double- counting can arise, in which it is likely to be less transparent than in fixing an aggregate sentence. The common case is where a mitigating factor has been taken into account so as to reduce the head sentence for an individual offence and also to reduce the non-parole period by a finding of special circumstances. Sentencing principles require the courts to avoid such outcomes. 6. It is implicit in the foregoing that where putative individual sentences are recorded, as they must be, the sentencing exercising will be no more complicated where convictions result from a combination of pleas and verdicts; the putative individual sentences will be indicated in each case. Indeed, this leads to a further important factor relevant to the fixing of an aggregate sentence. Although recording individual putative sentences is intended as an exercise in transparency, indicating the degree of criminality assessed with respect to each offence, there is no fixed relationship between the putative individual sentences and the aggregate sentence. The exercise of fixing an aggregate sentence may not be aptly described as instinctive, but is certainly an exercise in evaluative judgment. If it were necessary to determine the length, degree of accumulation and possible reduction for totality of the individual sentences, the exercise would (and does) involve a degree of artificiality and complexity. It is to avoid that exercise that aggregate sentencing is available. In one sense, the adoption of an aggregate sentence reduces the transparency of the calculation, but without significant cost, because the imposition of several individual sentences will not lend itself to precise or logical explanation. 7. Turning to s 22, there are both textual and purposive considerations to be addressed. First, the textual issue noted above relied on the expression, "an offence". In one sense, that language is logical; a plea can be entered only to an individual offence and the discount must be assessed according to the circumstances of each plea. On the other hand, the general principle of interpretation assumes that the singular form includes the plural. [38] If one sentence is to be imposed for several offences, multiple individual discounts can be taken into account, just as multiple putative individual sentences are to be reflected in a single sentence. 8. To apply the purposive approach, in accordance with s 33 of the Interpretation Act, it is necessary to identify the purpose (or purposes) of the relevant elements in the statutory scheme. The purpose of s 22 is to provide an incentive to offenders to enter a plea, rather than forcing the prosecution to prove its case at a trial. This provides a utilitarian benefit to law enforcement, as does the assistance in the prevention, detection or investigation of other offences, which is also rewarded, pursuant to s 23 of the Sentencing Procedure Act. As the Court recognised in R v Thomson and Houlton, [39] this effect is maximised where the discount is quantified and the discount given is expressly indicated and recorded. [40] 9. It may be argued that this beneficial purpose is served by the requirement in s 53A to record discounts on the putative individual sentences. That is partly true, but partly not. The lack of transparency involved in translating a number of individual sentences into an aggregate sentence means that the offender will not actually know how the discounts have been applied in reaching the aggregate sentence, unless it is expressly stated. The element of transparency required by s 22(2) is facilitated by requiring that the aggregate sentence be explained in terms of an undiscounted starting point and the actual sentence, after applying a specified discount. 10. There being no express disengagement of the operation of Pt 3 generally with respect to aggregate sentences (for example by way of qualification in s 4(3)), nor any express provision in either s 53A or s 22 which precludes s 22 applying to the determination of an aggregate sentence, it is arguable that the choice envisaged by s 33 of the Interpretation Act does not arise. However, acknowledging that there are arguments which would tend against the engagement of s 22, although they are not as persuasive on consideration as might appear at first blush, the mandate of s 33 of the Interpretation Act requires that the three sections be construed in a manner which promotes the purpose and object of s 22. Accordingly, as a matter of construction, s 22 is engaged in relation to the determination of an aggregate sentence. 11. Against this view, it may be said that there is caselaw which has assumed a contrary construction. That is true, although there is also caselaw which assumes the construction preferred above. However, none of the cases actually addresses the issue of statutory interpretation. Many accepted concessions by counsel and others simply adopted dicta which had not been challenged. In the present case, the issue as to the application of s 22 in determining an aggregate sentence was expressly raised by counsel for the applicant in the course of her submissions. (b) case law 1. Having regard to this conclusion, it is desirable to say something about the cases in this Court. It is convenient to start with the latest authority, Elsaj v R, [41] where it was said: [42] "The opening proposition of the applicant to the effect that 'a starting point of 22 years imprisonment is manifestly excessive in this case' is wrong and unhelpful. Section 53A of the Crimes (Sentencing Procedure) Act 1999 (NSW) makes it clear that discounts for a guilty plea are to be applied to the indicative sentences, not the aggregate sentence." The reasons referred to R v Nykolyn, [43] Subramaniam v R, [44] JM v R [45] and R v Cahill, [46] though without pinpoint references. There are also other cases to which Elsaj did not refer. 1. Taking the cases chronologically, a pattern emerges. 1. Nykolyn – there is no reference to the proposition that s 22(1) does not apply to an aggregate sentence. 2. Subramaniam – sentencing is discussed at [27]-[29], but again there is no reference to this issue. 3. SHR v R [47] – there is reference to indicative sentencing at [40], but no reference to the present point. 4. Stoeski v Regina [48] – the reasoning accepted at [33]-[34] that the s 22 discount does apply to an aggregate sentence. 5. JL v R [49] – accepted that the discount for a plea does apply to an aggregate sentence. 6. Khawaja v R [50] – This case considered whether the sentencing judge had adequately complied with the requirement to indicate putative individual sentences, it being unclear whether the discount for a plea had been applied to the putative individual sentences or to the aggregate sentence. There was no statement that the judge would have erred if he had applied the discount to the aggregate sentence; the issue was simply not discussed. 7. JM v R [51] – This case, frequently cited in relation to aggregate sentencing, set out a number of "principles" to be applied. Many of these principles were obiter, but usefully collected from earlier cases. Some involved general statements made without reference to specific statutory provisions. The relevant passage occurs in [39], at principle 3, which was not dealing with the question of discounting an aggregate sentence at all, but with the need to discount putative individual sentences, a proposition which is uncontroversial. The passage continued under the same heading: "In JL … at [54] it was said by way of conclusion in an appeal against the asserted severity of a sentence that 'The starting point for the aggregate sentence of 24 years before the allowance of a discount of 25 per cent to reflect the utilitarian value of the early pleas of guilty was not excessive'. This must be understood as a broad assessment within the conclusion rather than indicating that it is the aggregate sentence to which the discount should be applied. Stoeski … is anomalous in that at [33]-[34] it rejected a complaint that the sentencing judge had not discounted the aggregate sentence for the plea of guilty rather than rejecting the assertion that the discount applied to the aggregate sentence at all." This reasoning assumed that s 22 does not apply to an aggregate sentence, although at that date there was no authority which supported the assumption. It did not, as suggested in Elsaj, resolve any doubt which existed at that time. It was a passing observation which dismissed the two authorities which were to the contrary without any explanation as to why they were wrong. The fact that two members of the Court had also sat in Stoeski and one in JL suggests that this issue was not fully reconsidered. 1. Cahill – this case did involve an aggregate sentence which was discounted for the plea of guilty. [52] The reasons stated: [53] "It was erroneous to select an aggregate sentence and then apply a 25% discount for the Respondent's pleas of guilty to that sentence. The purpose of s 53A should be kept in mind. In the course of a valuable summary of principles surrounding aggregate sentencing, RA Hulme J … said in JM …: 'Section 53A was introduced in order to ameliorate the difficulties of applying the decision in Pearce … in sentencing for multiple offences: R v Nykolyn …. It offers the benefit when sentencing for multiple offences of obviating the need to engage in the laborious and sometimes complicated task of creating a "cascading or 'stairway' sentencing structure" when the principle of totality requires some accumulation of sentences….' It is not consistent with this principle to select the aggregate sentence and then apply the discount to the aggregate sentence." The passage quoted from JM does not directly support the proposition that one does not apply the s 22 discount to the aggregate sentence; nor is it clear how the passage explains why s 22(1) does not apply. 1. Glare v R [54] – In this case the error was failing to take into account pleas of guilty and provision of assistance in indicating the putative individual sentences; however, the statements in Cahill were repeated and, on resentencing, there was no indication that a discount was applied to the aggregate sentence, presumably because the putative individual sentences had in fact been discounted. (c) conclusion 1. As noted above, in no case has the question why a discount should not be applied to the aggregate sentence been squarely addressed. As Glare indicates, no injustice is likely to be done so long as the exercise of determining an aggregate sentence is undertaken by reference to the discounted individual putative sentences. However, it is likely that transparency would be improved, without practical difficulty, if s 22 (and s 23) were expressly addressed in determining the aggregate sentence. On the preferred construction of the Act, that is the course which should be taken. 2. BUTTON AND N ADAMS JJ: We have had the benefit of reading the judgment of Basten JA in draft. We respectfully agree with the orders proposed by his Honour for the reasons provided by his Honour insofar as they deal with the principles of parity in sentencing. We also agree that, because different factual findings were made about objective and subjective features pertaining to each offender, based on different evidence (including different agreed statements of facts) adduced in different proceedings on sentence, no erroneous disparity is demonstrated by the applicant in this case. 3. We respectfully disagree, however, with his Honour's conclusion that on a proper construction of ss 22 and 53A of the Crimes (Sentencing Procedure) Act 1999 (NSW) (the Act), in the context of aggregate sentencing, a discount for the utilitarian value of a plea of guilty should be applied both to the indicative sentences and to the aggregate sentence. (Basten JA uses the term "putative sentence", rather than "indicative sentence" but nothing turns on this difference in nomenclature.) 4. The statutory problem identified by Basten JA arises in this way. On the one hand, s 22(1) of the Act speaks of taking into account a plea of guilty, and possibly reducing the sentence on account of it, when "passing sentence". There is no doubt that when a judge imposes an aggregate sentence pursuant to s 53A of the Act it is the aggregate sentence that is actually imposed, not the indicative sentences. That suggests that a discount should be applied to the aggregate sentence that is actually passed. On the other hand, s 53A(2)(b) expressly requires that, when providing the indicative sentences, a sentencing judge is required to take into account all of the matters that are relevant under Part 3 of the Act. That part of the Act includes s 22, thereby calling for the application of discounts to the indicative sentences. 5. The practical consequence of the construction advanced by Basten JA is to require an explicit discount to be applied not only to each indicative sentence, but also to the ultimate aggregate sentence. 6. We have been unable to find any decision of this court since JM v R [2014] NSWCCA 297 (at [39]) that supports Basten JA's position on this issue. On the contrary, as Hoeben CJ at CL (Bathurst CJ and McCallum J agreeing) recently observed in Elsaj v R [2017] NSWCCA 124 at [56]: "Section 53A of the Crimes (Sentencing Procedure) Act 1999 (NSW) makes it clear that discounts for a guilty plea are to be applied to the indicative sentences, not the aggregate sentence. To the extent that there is any doubt on that issue, it has been resolved by such cases as R v Nykolyn [2012] NSWCCA 219, Subramaniam v R [2013] NSWCCA 159, JM v R [2014] NSWCCA 297 and R v Cahill [2015] NSWCCA 53." 1. The basis upon which Basten JA concludes that these decisions may be wrong, insofar as they concern the application of the discount for a plea of guilty to an aggregate sentence, is that none of them expressly consider s 22 of the Act in its terms nor how it is that ss 22 and 53A(2)(b) of the Act are to be read together harmoniously. His Honour has provided a detailed analysis of both the legislation and the relevant decisions and concluded that such a consideration leads to a different result. 2. We accept that no previous decision of this Court has expressly considered how it is that ss 22 and 53A(2)(b) of the Act are to be read together. Despite this, all of the decisions of this Court since JM v R are to the effect that a discount must be applied to the starting point of each indicative sentence and that there should be no explicit discount applied to the aggregate sentence. We believe that many judges of this Court have concurred in that analysis, and no doubt many hundreds of aggregate sentences have been imposed in accordance with it. We feel bound by that line of authority unless we consider those decisions to be "plainly wrong" (adopting the test posited by Heydon J in Green v The Queen (2011) 244 CLR 462; [2011] HCA 49 at 490-491 [84]-[85] and applied by this Court in Hampton v R [2014] NSWCCA 131 at [32]). 3. We have undertaken the exercise of considering how it is that ss 22 and 53A(2)(b) of the Act are to be read together harmoniously and have concluded that a proper construction of s 22 of the Act does not require the two-stage discounting procedure suggested by Basten JA when imposing an aggregate sentence. That is, we do not consider that the approach consistently taken by this Court since JM v R is plainly wrong. We take that view for the following reasons. 4. First, s 53A(2)(b) of the Act unambiguously calls for the application of a discount for a plea of guilty to the indicative sentence. There is no express provision in s 53A of the Act excluding the application of s 22 to the process of arriving at the indicative sentences. It is at this stage that the discount is to be applied. All three members of the Court agree on this point. 5. Secondly, the terms of s 22(1) of the Act require the court to "take into account" the plea of guilty when "passing sentence for an offence". We think that it is at least arguable that the imposition of an aggregate sentence founded upon one or more indicative sentences that reflect a discount for a plea of guilty complies with the section. This is so because the fact that the court has already taken pleas of guilty into account in fixing the indicative sentences means that those pleas are notionally taken into account in the aggregate sentence. In other words, in sentences where s 22 of the Act applies, the discount is "taken into account" indirectly when an aggregate sentence is imposed and directly in all other cases. 6. Thirdly, a construction that only requires the discount to be applied to the indicative sentences is consistent with the purpose of the aggregate sentencing regime. We respectfully agree with Basten JA that, applying a purposive approach in accordance with s 33 of the Interpretation Act 1987 (NSW), the Act in general and the statutory regime of aggregate sentencing in particular must be construed in accordance with their purposes. The purpose of s 22 of the Act is to encourage offenders to plead guilty. The incentive for doing so is that a discount will be applied to the sentence that would have otherwise been imposed. The purpose of s 22 is achieved if an offender can see that such a discount has been applied. 7. We consider that it is also relevant to bear in mind, however, that the purpose of the new regime of aggregate sentencing introduced in 2011 was to relieve judges who are sentencing for multiple matters from the complicated mathematical task of creating "cascading" sentence structures made up of many "overlapping" sentences, each with its own commencement and expiry date. (We will use the term "traditional sentencing" to describe sentencing structures that involve the fixing of a separate sentence, with its own commencement and expiry dates, for each offence.) Construing the sections together with both of these purposes in mind tends towards the construction that we advance. 8. Fourthly, it is not difficult to posit circumstances in which an explicit discount to be applied to the aggregate sentence will give rise to complications. Consider the aggregate sentence for two offences to be imposed on an offender who pleaded guilty at the earliest possible stage to one offence, but who fought and lost a trial with regard to the other. Is the discount to be applied to the aggregate sentence to reflect simply the average of the discounts of 25% and 0%? Does the relative seriousness of each offence have any role to play? Does the length of each indicative sentence have any role to play? 9. Separately, consider an offender who is to be sentenced for 10 separate offences by a District Court judge. He or she pleaded guilty to six of them when first charged before a magistrate the day after arrest. He or she pleaded guilty to three of them when first arraigned in the District Court. He or she pleaded guilty to the remaining one at 11 am on the morning of his or her trial. The discounts that should be applied to each individual indicative sentence can be readily determined. But what is the discount that should be applied to the overarching aggregate sentence? These complications argue against a construction that would lead to a requirement of discounting both the indicative sentences and the aggregate sentence. 10. Fifthly, as we understand it, the procedure suggested by Basten JA is to apply the specific discount (correlating with the timing of each plea) to each of the indicative sentences and then identify the rolled-up discount again when passing the aggregate sentence. That is, in the scenarios identified above at [82] – [83], the discounts ranging from 25% through to 10% or less would all be averaged out in some way. It is to be observed that, in Bao v R [2016] NSWCCA 16, this Court (Hoeben CJ at CL, with whom Bathurst CJ and RS Hulme AJ agreed) allowed a ground of appeal in which it was argued that a sentencing judge had "erred in combining and/or averaging the applicable discount referable to the applicant's pleas of guilty for the offences in question". 11. In Bao v R, the Crown agreed that, where there were multiple offences and pleas at different times, the utilitarian value of each plea needed to be separately considered with respect to each offence (R v SY and Anor [2003] NSWCCA 291; R v Borkowski [2009] NSWCCA 102; 195 A Crim R 1 at [32].) Hoeben CJ at CL found at [44] that the sentencing judge was in error in applying an average discount to the indicative sentences and that the correct procedure was to apply the appropriate discount to each of the indicative sentences. His Honour went on to identify the correct procedure to be that set out by Hamill J (Leeming JA and Fagan J agreeing) in Glare v R [2015] NSWCCA 194 at [12] – [15]. The Chief Judge then observed at [45] that averaging out the discounts "may well have affected the aggregate sentence to the detriment of the applicant". 12. Although Bao v R addressed a different complaint, it nonetheless seems to us that any attempt to "average out" the individual discounts to be applied to the aggregate sentence is problematic. It runs the risk of arriving at an "aggregate discount" that does not adequately reflect the individual discounts and thus leaves open a ground of appeal that the rolled up figure does not properly reflect the individual identifiable discounts on the indicative sentences. 13. Sixthly, as stated above, the purpose of the introduction of the regime of aggregate sentencing was to simplify the role of the sentencing judge when sentencing for multiple offences. Requiring a sentencing judge who is imposing an aggregate sentence to arrive at an arithmetically sound average of the various individual sentences is not consistent with this purpose. Instead, it adds a further layer of complexity to the already difficult task of sentencing offenders for multiple offences in the busy criminal lists in the District and Local Courts. 14. Seventhly, we accept that the purpose of discounts is to provide an incentive to plead guilty. It is true that the application of individual discounts to indicative sentences is less transparent than applying a further global discount to the aggregate sentence. But the general lack of transparency attaching to aggregate sentencing is the inevitable by-product of the fact that, as opposed to in traditional sentencing, one is not possessed of a sentencing structure that allows one to see the degree of accumulation, concurrence, or a combination thereof, between the individual sentences. That is because, of course, one does not have commencement and expiry dates that permit that analysis. 15. In the same vein, we accept that it is perhaps easier to analyse, in traditional sentencing, the degree to which discounts upon individual sentences inform the resulting total head sentence. As we have said, however, that lack of transparency is inherent in aggregate sentencing, the purpose of which was to simplify the task of sentencing for multiple offences. 16. Eighthly, as we have said, we do not believe that the purpose of the regime of aggregate sentences was to create a substantive change to sentencing law. Before the regime commenced, there was no suggestion that a sentencing judge would apply a discount to the total head sentence comprised of a number of individual discounted sentences, each with its own commencement date and expiry date. Even now, when a judge sentencing for multiple matters decides not to impose an aggregate sentence, each individual sentence actually imposed is discounted; thereafter, by way of the principles of totality (expressed by cumulation and concurrence, or a combination thereof), those individual discounts inform the length of the total head sentence. Nevertheless, it is not the case that the total head sentence is explicitly discounted in traditional sentencing. 17. In other words, in our view, to adopt the approach proposed by his Honour would constitute a substantive change to the law of sentencing. That would not be consistent with our understanding of the purpose of the regime of aggregate sentencing. 18. Finally, we do not believe that our approach occasions any injustice to offenders, and, in particular, to offenders who facilitate the administration of justice by entering pleas of guilty. That is because, as we have said, it is almost inevitable that the shortening of indicative sentences (by way of discounts being applied to them) will lead to an aggregate sentence that is itself shortened as a result. We say that because, speaking generally, and other things being equal, six undiscounted indicative head sentences of four years will lead to a longer aggregate sentence than six indicative sentences of four years discounted by 25% for an early plea of guilty to each offence. In other words, just as, in traditional sentencing, "cascading" discounted individual sentences will have earlier expiry dates, and will, by the application of the principle of totality, result in a shorter total head sentence (and a shorter total non-parole period), so too the same phenomenon surely occurs in the context of aggregate sentencing. 19. To express the above proposition in another way, it could be that our difference with Basten JA is a difference of route, not destination. By that we mean, whilst we do not accept that an explicit discount is to be applied to an aggregate head sentence, we do accept that the discounting of indicative sentences can be expected to lead to that outcome, in terms of the actual length of the aggregate sentence. 20. It is for the foregoing reasons that we respectfully are unable to agree with Basten JA that the correct approach to an aggregate sentence to which s 22 of the Act applies is to discount not only the indicative sentences but also the aggregate sentence. In our opinion, the approach proposed in JM v R, and adopted on many occasions by this Court since then, is correct. We are of the view that sentencing judges should continue to apply that approach when imposing an aggregate sentence, as should this Court when imposing an aggregate sentence on resentence in accordance with s 6(3) of the Criminal Appeal Act 1912 (NSW). 21. Furthermore, although it is true that the question under consideration arose during oral argument at the hearing of this appeal, we note that it was by no means fully argued, and formed no part of the written submissions of the parties. In these circumstances, we are even more reluctant to depart from the approach repeatedly adopted by this Court. 22. Finally, although we are not satisfied that the previous decisions of this Court concerning this issue are plainly wrong, that is not to say that the question is free from doubt, for the reasons raised by Basten JA. It seems to us that it would have been preferable had s 22 of the Act been amended at the same time that s 53A was enacted in order to make it clear that the discount is to be applied once and that it is to be applied only to the indicative sentences. If that was indeed the intention of Parliament, we respectfully suggest amendment of s 22 in order to free the question from doubt. ********** Endnotes 1. Sentencing Procedure Act, s 21A(1); Green v The Queen (2011) 244 CLR 462; [2011] HCA 49 at [19] (French CJ, Crennan and Kiefel JJ). 2. Green at [125]. 3. Green at [105]. 4. Lowe v The Queen (1984) 154 CLR 606 at 609 per Gibbs CJ; at 611-612 per Mason J; at 616 per Wilson J; at 617 per Brennan J; at 623 per Dawson J. 5. Lowe at 610 (Gibbs CJ); at 611-614 (Mason J); at 623-624 (Dawson J). 6. Lowe at 613. 7. Green at [31] (French CJ, Crennan and Kiefel JJ). 8. Postiglione v The Queen (1997) 189 CLR 295 at 323 (Gummow J); at 338 (Kirby J). 9. Lowe at 609. 10. Lowe at 617. 11. Lowe at 618-619. 12. Lowe at 614. 13. See fn 8 above. 14. Postiglione at 311 (McHugh J); 318 and 320 (Gummow J). 15. Postiglione at 314. 16. Postiglione at 320. 17. Postiglione at 300. 18. Postiglione at 300-301. 19. Postiglione at 342. 20. Postiglione at 343. 21. Judgment, p 19. 22. Judgment, p 20. 23. Judgment, p 17, see also p 20. 24. Judgment, p 23. 25. Judgment (Green), pp 21-22. 26. Judgment, pp 25-30. 27. Agreed facts, p 3. 28. Tcpt, 01/04/16, p 12(25). 29. Agreed facts, p 3. 30. Agreed facts (Green), p 3. 31. Tcpt, p 13(12)-(27). 32. Tcpt, p 14 (27). 33. Tcpt, p 27 (20)-(35). 34. Judgment (Green), p 6. 35. Judgment (Green), pp 6-7. 36. Judgment on sentence, 04/04/16, p 21. 37. Crimes (Sentencing Procedure) Amendment Act 2010 (NSW), Sch 2 [14]. 38. Interpretation Act 1987 (NSW), s 8(b). 39. (2000) 49 NSWLR 383; [2000] NSWCCA 309 at [160] (Spigelman CJ). 40. Sentencing Procedure Act, s 22(2). 41. [2017] NSWCCA 124 (Hoeben CJ at CL, Bathurst CJ and McCallum J agreeing). 42. Elsaj at [56]. 43. [2012] NSWCCA 219. 44. [2013] NSWCCA 159. 45. [2014] NSWCCA 297. 46. [2015] NSWCCA 53. 47. [2014] NSWCCA 94. 48. [2014] NSWCCA 161 (Adamson J, Hoeben CJ at CL and Bellew J agreeing). 49. [2014] NSWCCA 130 at [54] (McCallum J, Hoeben CJ at CL and Harrison J agreeing). 50. [2014] NSWCCA 80 (R S Hulme AJ, Leeming JA and Button J agreeing). 51. See fn 45. 52. Cahill at [70]. 53. Cahill at [109]-[110] (Johnson J, Leeming JA and Schmidt J agreeing). 54. [2015] NSWCCA 194 (Hamill J, Leeming JA and Fagan J agreeing). Amendments 03 August 2017 - [84] Correcting cross-references. [85] Amending "Hoeben CJ" to read "Hoeben CJ at CL". DISCLAIMER - Every effort has been made to comply with suppression orders or statutory provisions prohibiting publication that may apply to this judgment or decision. The onus remains on any person using material in the judgment or decision to ensure that the intended use of that material does not breach any such order or provision. Further enquiries may be directed to the Registry of the Court or Tribunal in which it was generated. Decision last updated: 03 August 2017