Concurrent and consecutive sentences

Part 4, Div 2 Crimes (Sentencing Procedure) Act 1999 (ss 55–60) contains provisions relating to the imposition of concurrent and consecutive sentences of imprisonment. It is convenient to explain here what DA Thomas first coined in his Principles of Sentencing, 2nd ed, 1979, Heinemann, London at p 56 as “the totality principle” (see A Ashworth, Sentencing and Criminal Justice, 4th ed, 2005, Cambridge University Press, New York at p 248).

[8-200] The principle of totality

Where a court sentences an offender for more than one offence, or sentences an offender serving an existing sentence, the aggregate or overall sentence must be “just and appropriate” to the totality of the offending behaviour. The High Court has quoted DA Thomas’ exposition of the common law principle (below) on at least three occasions, the most recent being Johnson v The Queen (2004) 78 ALJR 616 at [18]:

In Mill [Mill v The Queen (1988) 166 CLR 59 at 63] Wilson, Deane, Dawson, Toohey and Gaudron JJ adopted a statement from Thomas, Principles of Sentencing … at pp 56–57 [footnotes omitted]:

“The effect of the totality principle is to require a sentencer who has passed a series of sentences, each properly calculated in relation to the offence for which it is imposed and each properly made consecutive in accordance with the principles governing consecutive sentences, to review the aggregate sentence and consider whether the aggregate is ‘just and appropriate’. The principle has been stated many times in various forms: ‘when a number of offences are being dealt with and specific punishments in respect of them are being totted up to make a total, it is always necessary for the court to take a last look at the total just to see whether it looks wrong[’]; ‘when … cases of multiplicity of offences come before the court, the court must not content itself by doing the arithmetic and passing the sentence which the arithmetic produces. It must look at the totality of the criminal behaviour and ask itself what is the appropriate sentence for all the offences’.”

The passage from Thomas was also quoted in R H McL v The Queen (2000) 203 CLR 452 at [15] and R v Harris [2007] NSWCCA 130 at [44]. Street CJ’s description of the principle in R v Holder [1983] 3 NSWLR 245 is also commonly quoted, for example, in R v MMK [2006] NSWCCA 272 at [12]. Street CJ said at 260:

The principle of totality is a convenient phrase, descriptive of the significant practical consideration confronting a sentencing judge when sentencing for two or more offences. Not infrequently a straightforward arithmetical addition of sentences appropriate for each individual offence considered separately will arrive at an ultimate aggregate that exceeds what is called for in the whole of the circumstances. In such a situation the sentencing judge will evaluate, in a broad sense, the overall criminality involved in all of the offences and, having done so, will determine what, if any, downward adjustment is necessary, whether by telescoping or otherwise, in the aggregate sentences in order to achieve an appropriate relativity between the totality of the criminality and the totality of the sentences.

Totality both constrains and sets a lower limit

The Court of Criminal Appeal in R v MMK [2006] NSWCCA 272 at [11] said the principle of totality was “not-unrelated” to the principle of proportionality.

The task of the court is to ensure that the overall sentence is neither too harsh nor too lenient. Just as totality is applied to avoid a crushing sentence “… it is not to be disregarded for the converse purpose of assessing whether the overall effect of the sentences is sufficient …”: R v KM [2004] NSWCCA 65 at [55] cited with approval in Vaovasa v R [2007] NSWCCA 253 at [18]. The totality principle is routinely relied upon by the Crown in appeals against inadequacy of sentence. But mostly, the principle is invoked at first instance in the words of McHugh J in Postiglione v The Queen (1997) 189 CLR 295 at 308, whereby:

the Court … adjust[s] the prima facie length of the sentences downward in order to achieve an appropriate relativity between the totality of the criminality and the totality of the sentences.

Totality and public confidence in sentencing

In R v MAK [2006] NSWCCA 381 at [18], the court said the principle must be applied without a suggestion that a discount is given for multiple offences:

A sentencing court must, however, take care when applying the totality principle. Public confidence in the administration of justice requires the Court to avoid any suggestion that what is in effect being offered is some kind of a discount for multiple offending: R v Knight (2005) 155 A Crim R 252 at [112].

R v Harris [2007] NSWCCA 130 at [46] endorsed a statement of Sully J to similar effect in R v Wheeler [2000] NSWCCA 34 at [36].

[8-210] Applications of the totality principle

When a court is sentencing for multiple offences, and before it imposes the sentence for any one offence, it will have considered the outcome for all offences: R v JRD [2007] NSWCCA 55 at [33]. This approach ensures that the effective sentence reflects the overall criminality and that the individual sentences imposed conform to any statutory limitations that exist for specific sentencing options: R v JRD at [31], [33].

Where a sentence for an offence was reduced at first instance because of the totality principle on the basis of a premise, such as an existing sentence that no longer exists, a correction may be necessary so the sentence adequately reflects the criminality of the remaining offences, standing alone: Johnson v R [2017] NSWCCA 278 at [161]–[163]; R v Tolmie (1994) 72 A Crim R 416 at 418; see R v JDX [2017] NSWCCA 9 at [90]–[96]. See also Power to vary commencement of sentence at [8-270].

Totality and non-custodial sentences

The totality principle applies where a court imposes more than one non-custodial sentence, or a mixture of different non-custodial sentences, or imprisonment is imposed with an additional penalty or order: Camilleri’s Stock Feeds Pty Ltd v EPA (1993) 32 NSWLR 683 at 704; R v Chelmsford Crown Court; ex parte Birchell (1989) 11 Cr App R (S) 510 (fines); Winkler v Cameron (1981) 33 ALR 663 at 670 (fines and restitution orders); Hunter v White [2002] TASSC 72 at [9] (imprisonment and licence disqualification); and EPA v Barnes [2006] NSWCCA 246 at [50] (fines and an order for legal costs).

Totality and fines

Kirby P said in Camilleri’s Stock Feeds Pty Ltd v EPA (1993) 32 NSWLR 683 at 704:

The principle of totality is applicable where the penalty imposed is by way of fine: see R v Sgroi (1989) 40 A Crim R 197 at 203. However, it may be that the principle of totality may not have the same force in the case of the imposition of fines, as opposed to the imposition of imprisonment where it has a special operation: see R v Brown (1982) 5 A Crim R 404 at 407.

The passage was quoted with approval in EPA v Barnes [2006] NSWCCA 246 at [46]. Unlike terms of imprisonment, fines cannot be made “concurrent”. Each fine which is imposed must be paid separately.

The court in EPA v Barnes, above, at [50] suggested that if the sentencer believes that the totality principle requires an adjustment to the fines which may otherwise be appropriate, the amount of each fine should be altered by the approach taken by the first instance judge in Johnson v The Queen (2004) 78 ALJR 616 (discussed below), of reducing individual sentences and then aggregating each to determine a total fine amount.

[8-220] Totality and sentences of imprisonment

As to the application of the totality principle where a court is considering imposing intensive correction orders see [3-630].

A court which sentences an offender to more than one sentence of full-time imprisonment can utilise s 53A Crimes (Sentencing Procedure) Act 1999 and apply the principle of totality implicitly. See the discussion of the requirements for aggregate sentences at [7-505] and at [7-507] under the heading Application of Pearce v The Queen and the totality principle (propositions 1, 4 and 6). Alternatively the court can impose individual sentences (including a fixed terms/non-parole period and terms of sentence) with specific dates and apply the principle of totality explicitly.

The severity of a sentence of imprisonment is not purely linear

The court in R v MAK [2006] NSWCCA 381 at [15]–[18] identified at least two matters that are considered under the totality principle. The first is that:

The severity of a sentence is not simply the product of a linear relationship. That is to say severity may increase at a greater rate than an increase in the length of a sentence.

The court at [16] quoted R v Clinch (1994) 72 A Crim R 301 at 306–307 where Malcolm CJ said “a sentence of five years is more than five times as severe as a sentence of one year”. R v MAK and R v Clinch were referred to in Gore v R; Hunter v R [2010] NSWCCA 330 at [42]; Cavanagh v R [2009] NSWCCA 174 at [16]ff. However, sometimes very long sentences are required and it is not possible to determine whether inadequate weight has been given to what was said in Clinch until the court also reflects on other factors: Hampton v R [2010] NSWCCA 278 at [36]. For example, the effective non-parole period of two years was not beyond the available range in Einfeld v R [2010] NSWCCA 87, Latham J (RS Hulme J agreeing) at [201], Basten JA at [185]–[189] dissenting.

Imposition of a crushing sentence

The second matter, referred to by the court in R v MAK, above, at [17], is that the totality principle is designed to avoid a court imposing a “crushing sentence” or, as put by King CJ in R v Rossi (1988) 142 LSJS 451 at 453: “… where the total effect of the sentences merited by the individual crimes becomes so crushing as to call for the merciful intervention of the court by way of reducing the total effect.” The court in MAK explained the notion of a crushing sentence at [17]:

an extremely long total sentence may be “crushing” upon the offender in the sense that it will induce a feeling of hopelessness and destroy any expectation of a useful life after release. This effect both increases the severity of the sentence to be served and also destroys such prospects as there may be of rehabilitation and reform. Of course, in many cases of multiple offending, the offender may not be entitled to the element of mercy entailed in adopting such a constraint.

An assessment of whether a particular sentence is a “crushing sentence” must have regard to the offences committed, the maximum penalties, standard non-parole periods (if relevant) and all objective and subjective factors and principles concerning accumulation, concurrency and totality: Paxton v R [2011] NSWCCA 242 at [215]. The totality principle, including any necessity to avoid imposing a “crushing sentence”, is not a basis to avoid imposing a sentence that is “just and appropriate”. That a sentence may be “crushing” is but one matter taken into account in determining whether a particular sentence is beyond the range of sentences properly available: Hraichie v R [2022] NSWCCA 155 at [73]; Atai v R [2020] NSWCCA 302 at [132]; GS v R [2016] NSWCCA 266 at [50]–[51]. An extremely lengthy sentence would not necessarily be characterised as crushing if it reflects the total criminality of the offender’s conduct and would not be disturbed on appeal because the offender may feel crushed by it: Stanton v R [2017] NSWCCA 250 at [153]; ZA v R [2017] NSWCCA 132 at [76]–[85]. For young offenders a crushing sentence is one that is so long that the offender cannot conceive of enjoying a useful life after its expiration: Holliday v R [2013] ACTCA 31 at [61].

For a discussion on totality in the context of Commonwealth offences see Mohamed v The Queen [2022] VSCA 136 at [5]–[6] and Totality principle when previous sentence to be served: ss 16B, 19AD and 19AE in [16-050] Fixing non-parole periods and making recognizance release orders.

Statutory provisions for concurrent and consecutive sentences of imprisonment

Several provisions in the Act are relevant to sentencing exercises where more than one sentence of imprisonment is imposed. The provisions are technical in nature. The common law, discussed below, largely governs this area of the law.

Commencement date of sentences

Section 47(2)(a) provides that a court may direct that a sentence of imprisonment commence on a day prior to the day on which it is imposed. Section 47(2)(b) also provides that a court may direct that a sentence of imprisonment commence on a day occurring after the day on which the sentence is imposed but only if the sentence is to be served consecutively (or partly concurrently and partly consecutively) with some other sentence of imprisonment. A direction “must not” be later than the day following the earliest day on which it appears to the court that the offender will become eligible to be released from custody, or will become eligible to be released on parole: s 47(4). On the issue of backdating and forward dating sentences, see [12-500] Counting pre-sentence custody and [7-547] Forward dating sentences of imprisonment, respectively.

Multiple sentences of imprisonment

Section 53 provides:

(1) 

When a court imposes a sentence of imprisonment on an offender in relation to more than one offence, the court must (unless imposing an aggregate sentence of imprisonment in accordance with section 53A) comply with the requirements of this Division by imposing a separate sentence in relation to each offence.

(2) 

The term, and any non-parole period, set under this Division in relation to a 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.

Section 55 provides that in the absence of a direction where more than one sentence of imprisonment is imposed, or where the offender is subject to another sentence of imprisonment that is yet to expire, the sentence “is to be served concurrently”. Section 55 provides:

(1) 

In the absence of a direction under this section, a sentence of imprisonment imposed on an offender:

(a) 

who, when being sentenced, is subject to another sentence of imprisonment that is yet to expire, or

(b) 

in respect of whom another sentence of imprisonment has been imposed in the same proceedings,

is to be served concurrently with the other sentence of imprisonment and any further sentence of imprisonment that is yet to commence.

(2) 

The court imposing the sentence of imprisonment may instead direct that the sentence is to be served consecutively (or partly concurrently and partly consecutively) with the other sentence of imprisonment or, if there is a further sentence of imprisonment that is yet to commence, with the further sentence of imprisonment.

(3) 

A direction under this section has effect according to its terms.

Rather than create a presumption in favour of concurrency, s 55 appears to be directed to ensuring that, if the judge does not specifically address the issue, the default position is that the sentences are to be served concurrently: Yeung v R [2018] NSWCCA 52 at [46]. Section 55(1) does not require a specific direction. A direction is implicit in fixing the relevant commencement date and any more formalistic approach is not required: Yeung v R at [48].

Section 55(5) provides that s 55 does not apply to a sentence of imprisonment imposed on an offender in relation to an offence involving an assault, or any other offence against the person, committed by the offender while a convicted inmate of a correctional centre, or against a juvenile justice officer committed by the offender while a person subject to control, or a sentence of imprisonment imposed on an offender in relation to an offence involving an escape from lawful custody committed by the offender while an inmate of a correctional centre (whether or not the escape was from a correctional centre).

Smart AJ identified some practical problems that arise from the language used in s 55 in R v Killick [2002] NSWCCA 1 at [68]–[79].

[8-230] Structuring sentences of imprisonment and totality

It has been said that express legislative provisions apart (such as s 57 escape, see below) questions of concurrence or accumulation are a discretionary matter for the sentencing judge (R v Hammoud [2000] NSWCCA 540 at [7]; R v Scott [2005] NSWCCA 152 at [31]; LG v R [2012] NSWCCA 249 at [24]) and that in determining appropriate sentences:

Judges of first instance should be allowed as much flexibility in sentencing as is consonant with consistency of approach and as accords with the statutory regime under which the sentencing is effected. Johnson v The Queen (2004) 78 ALJR 616 at [26].

However, the court in R v MMK [2006] NSWCCA 272 at [13] made clear that “the discretion is generally circumscribed by a proper application of the principle of totality”. The court said at [11]:

It is the application of the totality principle that will generally determine the extent to which a particular sentence is to be served concurrently or cumulatively with an existing sentence in accordance with statements of the High Court as to the operation of the principle in Mill v The Queen (1988) 166 CLR 59; Pearce v The Queen (1998) 194 CLR 610 and Johnson v The Queen (2004) 78 ALJR 616.

Statements such as those in R v Hammoud [2000] NSWCCA 540 at [7] to the effect that questions of concurrence or accumulation are a discretionary matter for the sentencing judge “have to be read subject to what is required in a particular case to reflect the totality of the criminality before the Court”: R v Merrin [2007] NSWCCA 255 per Howie J at [36].

The following discussion sets out the common law position before aggregate sentences were introduced, as to which see ss 44(2A), (2C) and 53A discussed at [7-500]ff.

The “orthodox method” of setting sentences for each offence before considering the issues of concurrency or cumulation

The discussion will return to the issue of whether a sentence of imprisonment in a particular case ought to be served concurrently or made consecutive. The High Court has suggested specific approaches to setting sentences of imprisonment for multiple offences before issues of concurrency or cumulation are considered. In Mill v The Queen (1988) 166 CLR 59 the High Court suggested two approaches at 63:

Where the [totality] principle falls to be applied in relation to sentences of imprisonment imposed by a single sentencing court, an appropriate result may be achieved either by making sentences wholly or partially concurrent or by lowering the individual sentences below what would otherwise be appropriate in order to reflect the fact that a number of sentences are being imposed. Where practicable, the former is to be preferred [emphasis added].

The High Court in Johnson v The Queen (2004) 78 ALJR 616 [26] said that Mill v The Queen:

expresses a preference for what should be regarded as the orthodox, but not necessarily immutable, practice of fixing a sentence for each offence and aggregating them before taking the next step of determining concurrency [emphasis added].

In Pearce v The Queen (1998) 194 CLR 610 McHugh, Hayne and Callinan JJ said at [45]:

A judge sentencing an offender for more than one offence must fix an appropriate sentence for each offence and then consider questions of cumulation or concurrence, as well, of course, as questions of totality [emphasis added].

The court clarified in Johnson v The Queen at [26] that the approach suggested in Pearce v The Queen did not overrule the second method referred to in Mill v The Queen at 63 of “lowering the individual sentences below what would otherwise be appropriate” [emphasis added]:

Pearce does not decree that a sentencing judge may never lower each sentence and then aggregate them for determining the time to be served. To do that, is not to do what the joint judgment in Pearce holds to be undesirable, that is, to have regard only to the total effective sentence to be imposed on an offender [emphasis added].

Since Pearce v The Queen, the Court of Criminal Appeal has made clear that it is impermissible to impose a sentence for one offence and then increase it in order to encompass the criminality of other offences before the Court: R v Merrin [2007] NSWCCA 255 at [37]. Secondly, where an offender stands for sentence for multiple offences it is a clear error “… to have regard only to the total effective sentence to be imposed on an offender”: Johnson v The Queen at [26].

Brownie JA said in R v O’Connell [2005] NSWCCA 265 at [30] that the strict application of the approach suggested in Pearce may present a practical problem where individual offences, if considered individually, do not warrant a prison sentence.

Should a sentence of imprisonment be served concurrently or consecutively?

A sentence should not be concurrent “simply because of the similarity of the conduct or because it may be seen as part of the one course of criminal conduct … [t]he question to be asked is, can the sentence for one offence encompass the criminality of all the offences?”: R v Jarrold [2010] NSWCCA 69 per Howie J at [56], cited with approval in Franklin v R [2013] NSWCCA 122 at [44] and MPB v R [2013] NSWCCA 213 at [134].

In R v XX [2009] NSWCCA 115 at [52], Hall J derived the following 11 propositions from the case law, principally from Cahyadi v R [2007] NSWCCA 1 and Nguyen v R [2007] NSWCCA 14:

There is no general rule that determines whether sentences ought to be imposed concurrently or consecutively: see Cahyadi v R (2007) 168 A Crim R 41 per Howie J at 47. However, a number of propositions relevant to the consideration of that issue may be derived from the case law. They include the following:

(1) 

It is well established that questions of accumulation are, subject to the application of established principle, discretionary. What is important is that, firstly, an appropriate sentence is imposed in respect of each offence; and, secondly, that the total sentence imposed properly reflects the totality of the criminality: R v Wilson [2005] NSWCCA 219 at [38] per Simpson, Barr and Latham JJ agreeing.

(2) 

In R v Weldon (2002) 136 A Crim R 55, Ipp JA at [48] stated that it is “not infrequent that, where the offences arise out of one criminal enterprise, concurrent sentences will be imposed” but his Honour observed that “this is not an inflexible rule” and “[t]he practice should not be followed where wholly concurrent sentences would fail to take account of differences in conduct”.

(3) 

The question as to whether sentences in respect of two or more offences committed in the course of a single episode or a criminal enterprise or on a particular day should be concurrent or at least partly accumulated is to be determined by the principle of totality and the relevant factors to be taken into account in the application of that principle. See observations in this respect of Howie J in Nguyen v R [2007] NSWCCA 14 at [12].

(4) 

In applying the principle of totality, the question to be posed is whether the sentence for one offence can comprehend and reflect the criminality of the other offence. See generally R v MMK (2006) 164 A Crim R 481 at [11] and [13], Cahyadi at [12] and [27] and Vaovasa v R (2007) 174 A Crim R 116.

(5) 

If the sentence for one offence can comprehend and reflect the criminality of the other, then the sentences ought to be concurrent, otherwise there is a risk that the combined sentences will exceed that which is warranted to reflect the totality of the two offences: Cayhadi per Howie J at [27].

(6) 

If not, the sentence should be at least partially cumulative otherwise there is a risk that the total sentence will fail to reflect the total criminality of the two offences. This is so regardless of whether the two offences represent two discrete acts of criminality or can be regarded as part of a single episode of criminality: Cayhadi per Howie J at [27].

(7) 

Whether the sentence for one offence can comprehend and reflect the criminality of the other calls for the identification and an evaluation of relevant factors pertaining to the offences. These will include the nature and seriousness of each offence.

(8) 

In cases involving assault with violence where the offences involve two or more attacks of considerable violence and are distinct and separate (eg, see R v Dunn (2004) 144 A Crim R 180 at [50]) or in cases where there are separate victims of the attacks as in Wilson, the closeness in time and proximity of the two offences will often not be determinative factors. See also R v KM [2004] NSWCCA 65. In Wilson, having regard to the purposes of sentencing set out in s 3A of the Crimes (Sentencing Procedure) Act, Simpson J observed at [38] that “… to fail to accumulate, at least partially, may well be seen as a failure to acknowledge the harm done to those individual victims …”

(9) 

Where two offences committed during the course of a single episode are of a completely different nature and each individually involved significant or extreme gravity, it is likely that some accumulation will be necessary to address the criminality of the two: Nguyen per Howie J at [13] .

(10) 

Possession of two different kinds of drugs may not be regarded as one episode of criminality in a case of “deemed” supply: Luu v R [2008] NSWCCA 285 at [32].

(11) 

The fact that the evidence of two offences (eg, documentary evidence or the presence of drugs) are located by police at or in the one place is not a relevant factor in favour of concurrent sentences … (Cahaydi at [26]).

Iskov v R [2011] NSWCCA 241 at [87]–[91] is an example of an application of Cahyadi v R. It was held that even though three offences had been committed against the same victim within a period of a few hours, the judge was required to make each sentence partly cumulative on the preceding sentence or sentences because the criminality in each offence could not be comprehended within the other offences.

Multiple victims and discrete offending usually require partly consecutive sentences

The following cases for dangerous driving, sexual assault, assault and wounding, break, enter and robbery are cited as examples. The cases hold that the fact that there is more than one victim will generally require an increase in the otherwise appropriate sentence than where only one victim was involved: Vaovasa v R [2007] NSWCCA 253 at [16]. Similarly a prudent measure of cumulation is necessary where the criminal conduct is capable of being described as discrete offending.

Dangerous driving cases

In R v Janceski [2005] NSWCCA 288, Hunt AJA at [23] explained the approach of sentencing for a single action aggravated by multiple victims:

separate sentences should usually be fixed which are made partly concurrent and partly cumulative, each such sentence being appropriate to the existence of only one victim and the aggregate of the sentences reflecting the fact that there are multiple victims resulting from the same action by the offender.

Further driving cases where the issue has been discussed include: R v Skrill [2002] NSWCCA 484 at [75]; R v Plumb [2003] NSWCCA 359 at [12] and cases listed at [19]; Richards v R [2006] NSWCCA 262. In the latter case it was said at [78]: “… failure to accumulate those sentences, at least partially, appears to have been a failure to acknowledge the harm done to the individual victims”. See also [18-400] Totality in Dangerous driving and navigation.

Sexual assault

Generally, relevant considerations include the number of victims and whether the offences committed against each occurred on separate occasions: Van der Baan v R [2012] NSWCCA 5 at [117]. Where sexual offences arise out of one event a court is required to identify a sentence appropriate for each separate act and some degree of accumulation is sometimes necessary to address additional criminality: Franklin v R [2013] NSWCCA 122 at [44]–[45]. It is open for a court to make each victim’s sentence wholly cumulative upon the non-parole period of another victim where the offences are committed on separate victims over an extended period: Magnuson v R [2013] NSWCCA 50 at [142]. It was an error in Nguyen v R [2007] NSWCCA 14 at [13] for the court to impose wholly concurrent sentences for the offences of armed robbery and sexual intercourse without consent in circumstances of aggravation which arose from the same incident. Similarly in R v Gorman [2002] NSWCCA 516 at [9], the judge erred by imposing wholly concurrent sentences for sexual offences arising from the same incident. Characterising the offences as “one episode of criminality” misapplied Pearce v The Queen and failed to have regard to the specific circumstances of each individual offence.

Further sexual assault cases where a judge has erred by imposing wholly concurrent sentences for discrete offending include R v Smith [2006] NSWCCA 353 at [17] and [23]; R v TWP [2006] NSWCCA 141 at [25]–[27]; R v BWS [2007] NSWCCA 59 at [16]–[17].

Where a court is required to sentence according to past (the late 1970s to early 1980s practices) it must be borne in mind that “the approach to questions of concurrence and cumulation was more lax, before the handing down of Pearce v The Queen”: Magnuson v R per Button J at [143]. This does not apply to child sexual offences. Section 25AA Crimes (Sentencing Procedure) Act 1999, which came into force on 31 August 2018, requires a court sentencing for such an offence to sentence the offender in accordance with sentencing patterns and practice at the time of sentencing, not at the time of the offence.

Assault and wounding offences

The judge in R v Dunn [2004] NSWCCA 41 erred by imposing concurrent sentences for two offences involving wounding committed in the course of a single extended criminal episode. Adams J expressed the view at [50]:

There is a distinct difference between assaulting one victim and assaulting two. Each was intentionally injured with the knife. The learned sentencing judge did not articulate his reasons for making the sentences wholly concurrent.

The judge erred in R v Nguyen [2013] NSWCCA 195 by imposing wholly concurrent sentences for both a wounding with intent to cause grievous bodily harm offence under s 33(1)(a) Crimes Act 1900 and manslaughter. Although there was short period of time between the offences, they were distinct offences caused by different bullets resulting in very different consequences: R v Nguyen at [81]. The nature and seriousness of the wounding offence was such that the sentence for manslaughter could not sufficiently comprehend and reflect the criminality involved in the wounding offence: R v Nguyen at [83].

Robbery

Where there are multiple counts it is incumbent on the court to consider the question of totality: R v Kelly [2010] NSWCCA 259. Imposing fixed terms for all but the most serious charge is “inappropriate in the context of serious offences such as robbery”: R v Kelly at [55]. The judge’s erroneous global approach caused her to underestimate the seriousness of the first (home invasion) offence: R v Kelly at [56]. In Vaovasa v R [2007] NSWCCA 253 at [19] the judge erred by imposing wholly concurrent sentences for three robbery in company offences upon the basis that the offences were committed against three victims and were part of one course of criminality of short duration.

Break, enter and steal

Totality will rarely, if ever, justify wholly concurrent sentences for a series of break enter offences: R v Merrin [2007] NSWCCA 255 at [38] citing R v Harris [2007] NSWCCA 130 at [38]–[42]. The judge in Harris erred by imposing wholly concurrent sentences for a “series of [break enter] offences”. The court held at [45] that the limiting or constraining function of the principle of totality:

will rarely if ever go so far as to justify wholly concurrent sentences for all of a series of offences such as those here. Subject to those limits, in general, sentences significantly cumulative should be imposed for separate serious offences of which those here are all examples.

Earlier at [40] the court said:

Making sentences wholly concurrent means that the second and subsequent effectively constitute no punishment and sends a clear message to those members of the criminal community who chose to live by breaking and entering and stealing or the like that once they have committed one or a few offences, they can continue offending with virtual impunity so far as sentences are concerned.

The court acknowledged the circumstances where wholly concurrent sentences may be justified for break enter offences at [43]:

Of course at times there will be good reason for complete concurrency. One is where some offences are little more than incidents of, or incidental to, others.

Fraud offences

In R v Hawkins [2013] NSWCCA 208 at [23]–[24], it was held that the individual sentences for the charges of defraud the Commonwealth and obtain financial advantage by deception may have been appropriate, however the concurrency of the sentences had the effect that the respondent received no punishment for six of the offences.

Offences in contravention of apprehended domestic violence orders

An offence committed in breach of an apprehended domestic violence order (ADVO), and an offence of breaching an ADVO, involves separate and distinct criminality. There is no duplicity in imposing distinct sentences for what are distinct offences: Suksa-Ngacharoen v R [2018] NSWCCA 142 at [131]. Conduct involving deliberate disobedience of a court order must be treated as serious, and should ordinarily be separately punished from any offence occurring at the same time: Suksa-Ngacharoen v R at [132].

Totality and existing sentences of imprisonment

The totality principle has been applied where an offender is serving an existing sentence and is sentenced by the second court a period after the first offence: Mill v The Queen (1988) 166 CLR 59 at 66; Choi v R [2007] NSWCCA 150 at [157]. The court in Mill at 66 said that in a case where the offences were committed in a short period across State borders the proper approach:

was to ask what would be likely to have been the effective head sentence imposed if the applicant had committed all three offences of armed robbery in one jurisdiction and had been sentenced at one time.

The principle is applied not just to the non-parole period but to the head sentence: Mill at 67.

If the criminality of offences previously committed is great there will be very little room left for a further penalty to be imposed: R v MMK [2006] NSWCCA 272 at [14].

Offences committed under both state and federal law

Section 16B Crimes Act 1914 (Cth) gives statutory expression to the principle where an offender is sentenced for Commonwealth offences: Postiglione v The Queen (1997) 189 CLR 295 at 308. Section 16B provides that the court is to have regard to:

(a) 

any sentence already imposed on the person by the court or another court for any other federal offence or for any State or Territory offence, being a sentence that the person has not served; and

(b) 

any sentence that the person is liable to serve because of the revocation of a parole order made, or licence granted, under this Part or under a law of a State or Territory.

Separate indictments

There are no special rules in relation to totality which apply where a judge sentences an offender for charges on more than one indictment: R v Finnie [2002] NSWCCA 533 at [57]–[58]. Concurrent sentences should not have been imposed in R v Finnie because the separate indictments were referable to different and separate episodes of criminal activity and involved different modus operandi and different victims.

Totality and overlapping charges

An offender should not be punished twice for common elements between offences. In Pearce v The Queen (1998) 194 CLR 610, McHugh, Hayne and Callinan JJ said at [40]:

To the extent to which two offences of which an offender stands convicted contain common elements, it would be wrong to punish that offender twice for the commission of the elements that are common. No doubt that general principle must yield to any contrary legislative intention, but the punishment to be exacted should reflect what an offender has done; it should not be affected by the way in which the boundaries of particular offences are drawn. Often those boundaries will be drawn in a way that means that offences overlap. To punish an offender twice if conduct falls in that area of overlap would be to punish offenders according to the accidents of legislative history, rather than according to their just deserts.

The court said at [41] it “… need not decide whether this result is properly to be characterised as good sentencing practice or as a positive rule of law”.

The principle was explained in Nahlous v R [2010] NSWCCA 58 at [17]:

a person can by the one act commit two offences and, where the two offences address different aspects of the criminal conduct, there is nothing wrong with prosecuting the two offences or, subject to the principle of totality, with imposing separate sentences for the two offences.

In Pearce v The Queen, the overlapping charges were maliciously inflict grievous bodily harm with intent to do grievous bodily harm and break and enter a dwelling house and while therein inflict grievous bodily harm. The court concluded at [49]: “ … the individual sentences imposed on counts 9 and 10 were flawed because they doubly punished the appellant for a single act, namely, the infliction of grievous bodily harm”.

The double punishment principle referred to in Pearce v The Queen was applied in R v Hilton [2005] NSWCCA 317 where the applicant was charged with 11 counts of obtaining money from child prostitution under s 91E(1) Crimes Act and eight counts of using premises for child prostitution under s 91F(1). The court held that he was doubly punished for his conduct.

The practice or rule is obviously not applicable where there is a specific statutory provision which prevents the Crown charging a person with two offences with different ingredients for the same conduct or where it would be oppressive to charge for the second offence. Section 25A(5) Drug Misuse and Trafficking Act 1985 provides that a person who has been convicted of an offence under s 25A (ongoing supply) is not liable to be convicted of an offence under s 25 (supply) on the same or substantially the same facts: Tran v R [2007] NSWCCA 140 at [12]. In Nahlous v R [2010] NSWCCA 58 at [17], the court held it was oppressive to charge for both the sale of the illegal decoders and also the receipt of the money as a result of the sale. The sale offence encompassed the criminality of possessing the proceeds of the sale. On the other hand, receiving stolen property (even where the stolen property happens to be drugs) is quite different from the act of criminality in possessing a drug for the purpose of sale: Hinchcliffe v R [2010] NSWCCA 306 at [27].

[8-240] Sentences for offences involving assault by convicted inmate

Section 56 sets out specific provisions for sentences of imprisonment imposed on an offender in relation to “an offence involving an assault, or any other offence against the person, committed by the offender while a convicted inmate of a correctional centre” (s 56(1)(a)) or “an offence involving an assault, or any other offence against the person, against a juvenile justice officer committed by the offender while a person subject to control” (s 56(1)(b)). The sentence of imprisonment “is to be served consecutively” (s 56(2)) or the court “may instead direct that the sentence is to be served concurrently (or partly concurrently and partly consecutively) with the other sentence of imprisonment and any further sentence of imprisonment that is yet to commence”: s 56(3). Such a direction may not be made for an offence involving an assault against a correctional officer or a juvenile justice officer unless the court is of the opinion that there are special circumstances justifying such a direction: s 56(3A).

If the court makes an order under s 56(3), that the second sentence is to be served concurrently or partly consecutively, the reasons for doing so have to be exposed: R v Hoskins [2004] NSWCCA 236 at [31]. In that case the effective sentence did not adequately reflect the seriousness of the crime and insufficient weight was given to general deterrence: R v Hoskins at [62]–[63] citing R v Fyffe [2002] NSWSC 751. See also R v Windle [2012] NSWCCA 222 at [56]. In Banks v R [2018] NSWCCA 41, the judge erred by wholly accumulating a sentence for recklessly wounding an inmate, on lengthy sentences already being served, resulting in an overall non-parole period of 14½ years that was 92% of the overall head sentence. It was in both the community and the applicant’s interests that a longer period than 15 months of supervision on parole be available: at [32]–[34].

Section 56 does not apply when an offender has been released on parole but remains in custody and bail refused for subsequent offences — such an offender is not a “convicted inmate of a correctional centre”: Hraichie v R [2022] NSWCCA 155 at [132]–[133], [136], [145] (note the offender was on parole by virtue of s 50 (rep) of the Crimes (Sentencing Procedure) Act which then required a court which had sentenced an offender to 3 years imprisonment or less to make a parole order directing their release at the end of the non-parole period).

Although s 56 only applies to “convicted inmates”, the policy objectives behind it have been applied in R v Jeremiah [2016] NSWCCA 241 where an offender committed an assault whilst on remand; in Tammer-Spence v R [2021] NSWCCA 90 (see [42], [45]–[46]), where the offender poured boiling water on his cell mate while serving the balance of parole for a sentence for armed robbery; and in Hraichie v R [2022] NSWCCA 155 (see at [148]) where the offender committed offences of aggravated kidnapping and assaulting an inmate while bail refused for other offences.

[8-250] Sentences for offences involving escape by inmate

Section 57 sets out specific provisions for sentences of imprisonment imposed on an offender in relation to an offence involving an escape from lawful custody committed by the offender while an inmate of a correctional centre. Part 6A Crimes Act 1900 sets out offences relating to escape from lawful custody. Section 310D provides for an offence for an inmate who escapes or attempts to escape from lawful custody. Where the court is sentencing an offender for an offence involving escape from lawful custody, the court must set the sentence for “non-escape” offences first so that “escape” offences will be cumulative on them. Section 57(1A) provides:

A sentence of imprisonment to which this section applies must be imposed after any other sentence of imprisonment that is imposed in the same proceedings.

Section 57(2) provides that where an offender is an “inmate of a correctional centre” and commits an offence “involving escape”, the sentence is to be served consecutively. See for example R v Mathieson [2002] NSWCCA 97 at [30]. The statutory requirement in s 57 was not mentioned or put into effect in Jinnette v R [2012] NSWCCA 217 at [90]–[96].

In R v Pham [2005] NSWCCA 94, quoted with approved in Jinnette v R, Wood CJ at CL, with whom Hislop and Johnson JJ agreed, said at [16]–[19]:

The offence of escape has been regarded by the courts as a serious offence, which potentially jeopardises the future of minimum security facilities and threatens the continued provision to prisoners of beneficial and humanitarian custodial arrangements and opportunities. It may lead to additional restrictions being placed upon their access to external medical treatment, and it may also impede the progress of rehabilitation for offenders with favourable prospects, if conditions of detention are strengthened, in order to prevent escapes.

These considerations were noted, for example, in R v Thomson NSWCCA 21 May 1986 where, in a case decided before enactment of the Sentencing Act 1989, Street CJ observed that the ordinary sentence for an unremarkable escape “could be expected to approximate two years” (at a time when the maximum penalty for the offence was imprisonment for seven years); and also in R v Mathieson [2002] NSWCCA 97 at [27].

Where the offender has remained at large for a very lengthy period or has used the opportunity of being at large to commit further offences, as was the case here, then the overall objective seriousness of his criminality is potentially increased: R v Plummer [2000] NSWCCA 363 at [34] and R v Josef Regina [2000] NSWCCA 100. The elements of both personal and general deterrence are also important, it being essential that prisoners understand that any offence of escape or attempted escape will result in a meaningful overall increase in their detention: R v Butler [2000] NSWCCA 525 at [18] and R v Smith [2004] NSWCCA 69. That this is so is also demonstrated by the fact that the maximum penalty prescribed for the offence has been increased from imprisonment for 7 years to imprisonment for 10 years.

It is also for that reason that the legislature enacted, by way of s 57(2) of the Crimes (Sentencing Procedure) Act 1999, a requirement for sentences for escape to be served consecutively upon any existing sentence that has yet to expire, or upon any other sentence that is imposed in the same proceedings.

R v Pham was complicated in so far as it involved a consideration of s 57(3), as well as s 47. The respondent escaped during the parole period of an existing sentence and was at large for a considerable period of time. The court held that there were two distinct purposes apparent from these provisions: the first was to ensure that the offence of escape attracted an actual and meaningful accumulation of sentence; the second was to avoid the existence of a possible hiatus in custody, which would arise if the offender was later released to parole for the existing sentence before the date fixed for commencement of the fresh sentence. It held that the commencement date of the new sentence was discretionary and governed by s 47 of the Act. The sentence was within the appropriate range but the starting date required adjustment in order to reflect an adequate period of additional punishment.

Section 254 Crimes (Administration of Sentences) Act 1999 supplements the operation of Pt 6A Crimes Act 1900, in that the section allows for sentences to be extended where an offender is unlawfully absent from custody. However, the section does not operate to prevent a person from being proceeded against and convicted of any offence arising out of an escape: s 254(4).

[8-260] Limitation on consecutive sentences imposed by Local Courts

Sections 267(2) and 268(2) Criminal Procedure Act 1986 provide that the maximum term of imprisonment that the Local Court may impose for an offence is, subject to the relevant section, 2 years or the maximum term of imprisonment provided by law for the offence, whichever is the shorter term. The former section applies to Table 1 offences and the latter to Table 2 offences.

Section 58 Crimes (Sentencing Procedure) Act 1999 sets numerical limitations on consecutive sentences imposed by the Local Court. Section 58 is a very technical provision and close attention must be given to the language of the section. Its language is a consequence of the troubled history which plagued its predecessor (s 444 Crimes Act 1900): see R v Clayton (1997) 42 NSWLR 268. It provides:

58 Limitation on consecutive sentences imposed by Local Courts

(1) 

A Local Court may not impose a new sentence of imprisonment to be served consecutively (or partly concurrently and partly consecutively) with an existing sentence of imprisonment if the date on which the new sentence would end is more than 5 years after the date on which the existing sentence (or, if more than one, the first of them) began.

(2) 

Any period for which an existing sentence has been extended under this or any other Act is to be disregarded for the purposes of this section.

(3) 

This section does not apply if:

(a) 

the new sentence relates to:

(i) 

an offence involving an escape from lawful custody, or

(ii) 

an offence involving an assault or other offence against the person, being an offence committed (while the offender was a convicted inmate) against a correctional officer or (while the offender was a person subject to control) against a juvenile justice officer, and

(b) 

either:

(i) 

the existing sentence (or, if more than one, any of them) was imposed by a court other than a Local Court or the Children’s Court, or

(ii) 

the existing sentence (or, if more than one, each of them) was imposed by a Local Court or the Children’s Court and the date on which the new sentence would end is not more than 5 years and 6 months after the date on which the existing sentence (or, if more than one, the first of them) began.

(4) 

In this section:

“existing sentence” means an unexpired sentence, and includes any expired sentence or unbroken sequence of expired sentences with which the unexpired sentence is being served consecutively (or partly concurrently and partly consecutively).

“sentence of imprisonment” includes an order referred to in section 33(1)(g) of the Children (Criminal Proceedings) Act 1987.

Section 58 empowers the Local Court to accumulate sentences up to five years within the prescribed limits outlined above. The operation of s 58 was considered in R v Perrin [2022] NSWCCA 170 where Wright J (Ward P and Harrison J agreeing) held that:

1. 

If there is no existing sentence, s 58 is not engaged: [80], [81].

2. 

Whether a sentence is “existing” or “unexpired” for the purposes of s 58(1) and (4) is determined on the date the new sentence is imposed: [46], [66], [80]; Stoneham v Director of Public Prosecutions (NSW) [2021] NSWSC 735 at [33].

3. 

The prohibition or limitation in s 58(1) relates directly to imposing a new sentence to be served consecutively or partly concurrently and partly consecutively with an existing sentence only if the expiry date of the new sentence is more than 5 years after the existing sentence commenced: [46], [71].

4. 

The practical effect of s 58 is to constrain to a greater or lesser extent the length of the new sentence: [71]–[77].

Accordingly, when the new sentence is imposed, the Local Court must determine:

  • whether there is an “existing sentence”, being an “unexpired sentence” that also includes any expired sentence or unbroken sequence of expired sentences with which the unexpired sentence “is being served” wholly or partly consecutively; and

  • whether the date on which the new sentence would end is more than 5 years after the date on which that “existing sentence” began: R v Perrin at [46].

Section 58 and aggregate sentences

Section 58 only applies to the imposition of a sentence which is to be served consecutively. As for aggregate sentences: see Aggregate sentences at [7-505].

Section 53B permits the Local Court to impose an aggregate sentence of up to five years. It does not alter the jurisdictional limit of two years for individual offences referred to above.

[8-270] Power to vary commencement of sentence

Section 59 provides:

59 Court may vary commencement of sentence on quashing or varying other sentence

(1) 

A court that quashes or varies a sentence of imprisonment imposed on a person (on appeal or otherwise) may vary the date of commencement of any other sentence that has been imposed on that person by that or any other court.

(2) 

If a person is subject to two or more sentences, this section applies to each of them.

(3) 

A court may vary a sentence under this section on its own initiative or on the application of a party to the proceedings on the quashing or variation of the other sentence.

(4) 

An appeal does not lie merely because the date of commencement of a sentence is varied under this section.

(5) 

The term of a sentence, or the non-parole period of a sentence, cannot be varied under this section.

The provision is designed to remedy a difficulty where the quashing of a sentence following a successful appeal, usually in the District Court or the Court of Criminal Appeal, leaves the appellant with a further sentence of imprisonment to commence on a specified date in the future. It was regarded as being both impractical and unjust to return a person to custody on a future date. Section 59 was amended by the Crimes Legislation Amendment Act 2003 to remove a reference to “consecutive” and enable the section to be applied to concurrent sentences and partially consecutive sentences: Allan v R (No 2) [2011] NSWCCA 27 at [13]. The power in s 59 is not limited to the scenario where the quashing or varying of a sentence will result in a hiatus for a further sentence of imprisonment which commences on a date in the future: Allan v R (No 2) at [18]. A court may vary the date of commencement of any other sentence that has been imposed on that person by any other court if by quashing the sentence(s) there is no change in an offender’s release date: Allan v R (No 2) at [19].

[8-280] Application of Division to interstate sentences of imprisonment

Part 4 Div 2 Crimes (Sentencing Procedure) Act 1999 applies to unexpired sentences passed outside NSW, or to be served within NSW, in the same way as it applies to unexpired sentences passed within NSW: s 60.