Reported Decision :

46 NSWLR 346
106 A Crim R 149

New South Wales
Court of Criminal Appeal
CITATION :REGINA v HENRY BARBER TRAN SILVER TSOUKATOS KYROGLOU JENKINS [1999] NSWCCA 111 revised - 18/05/99
FILE NUMBER(S) :CCA 60559/98; 60558/98; 60561/98; 60746/98; 60596/98; 60595/98; 60511/98
HEARING DATE(S) :18/03/99-19/03/99
JUDGMENT DATE :
12 May 1999

PARTIES :


Paul Anthony HENRY
Stephen Anthony BARBER
Hoai Vinh TRAN
Troy David SILVER
Theo TSOUKATOS
Bill KYROGLOU
John David JENKINS
JUDGMENT OF :Spigelman CJ at 1; Wood CJ at CL at 213; Newman J at 278; Hulme J at 279; Simpson J at 332
LOWER COURT JURISDICTION :District Court
LOWER COURT FILE NUMBER(S) : 97/31/0416; 98/31/0291; 98/31/0275; 98/21/2020; 98/31/0368; 98/21/0135; 98/21/0136; 98/61/0001
LOWER COURT JUDICIAL OFFICER :Johnston ADCJ, Sides DCJ, Prowse ADCJ, Cantrill ADCJ
COUNSEL : P J D Hamill (Henry)
J S Stratton (Barber)
P Byrne SC (Tran)
P M Strickland (Silver)
S J Odgers (Tsoukatos)
S J Odgers (Kyroglou)
T L Buddin SC (Jenkins)
P G Berman (Crown)
SOLICITORS : Rummery Stewart (Henry)
T A Murphy (Barber)
T A Murphy (Tran)
T A Murphy (Silver)
Konstan Lawyers (Tsoukatos)
Konstan Lawyers (Kyroglou)
T A Murphy (Jenkins)
R Gray (Crown)
CATCHWORDS :CRIMINAL LAW; sentencing; guideline judgment; appropriateness of promulgating guidelines; increased incidence of offence; inconsistency and systematic leniency; appropriateness of custodial sentence; common type of offence; starting point or sentencing range; role of drug addiction in sentencing exercise; CRIMINAL LAW; evidence; use of crime statistics; use of new material in sentencing appeal; "evidence" s12 Criminal Appeal Act 1912 (NSW), s29A Criminal Procedure Act 1986 (NSW); CRIMINAL LAW; particular offences; armed robbery; s97(1) and s97(2) Crimes Act 1900 (NSW)
ACTS CITED :Crimes Act 1900 (NSW)
Crimes Act 1914 (Cth)
Criminal Appeal Act 1912 (NSW)
Criminal Code 1899 (QLD)
Criminal Procedure Act 1986 (NSW)
Criminal Procedure Amendment (Sentencing Guidelines) Act 1998 (NSW)
Evidence Act 1995
Criminal Law Consolidation Act 1925 (SA)
Sentencing Act 1989 (NSW)
Sentencing Act 1995 (WA)
Statutes Amendment (Truth in Sentencing) Act 1994 (SA)
DECISION :




- 7 -
IN THE COURT OF
CRIMINAL APPEAL


60559/98
60558/98
60561/98
60746/98
60596/98
60595/98
60511/98


                                SPIGELMAN CJ
                                WOOD CJ at CL
                                NEWMAN J
                                HULME J
                                SIMPSON J


                                Wednesday 12 May 1999

                REGINA v Paul Anthony HENRY
                REGINA v Stephen Anthony BARBER
                REGINA v Hoai Vinh TRAN
                REGINA v Troy David SILVER
                REGINA v Theo TSOUKATOS
                REGINA v Bill KYROGLOU
                REGINA v John David JENKINS

Six Crown appeals and one severity appeal involving the offence of armed robbery under s97(1) (one involving the aggravated offence under s97(2)) of the Crimes Act 1900 (NSW) were heard together. The Crown submitted that it was appropriate to promulgate a guideline judgment with respect to this offence.

Held (By the Court):
Sentencing Guidelines: General
The purpose of a guideline judgment is to foster consistency in sentencing: Jurisic NSWCCA 12 October 1998; Lowe (1994) 154 CLR 606 at 610-611.

The appropriateness of an appellate court promulgating guidelines has been authoritatively established: Norbis v Norbis (1986) 161 CLR 513.

A guideline judgment on the subject of sentencing does not lay down anything in the nature of a rule. The failure to sentence in accordance with a guideline is not itself a ground of appeal: Norbis per Brennan J at 536, 537-538; Jurisic at pp15,16.

Consideration of the Canadian and South Australian approaches to sentencing guideline. McDonnell (1997) 114 CCC (3d) 4336; Police v Cadd (1997) 94 ACrimR 466; Bini (1994) 68 ALJR 859, referred to.

Use of New Material in Individual Cases
Consideration of the use of material before the appellate court that was not before the sentencing judge. The word “evidence” in s12 of the Criminal Appeal Act 1912 and s29A of the Criminal Procedure Act 1986 is confined to “matters going to the commission of the offence or the personal circumstances of the accused”: Beldan (1986) 21 ACrimR 159; Chanh Nghia Ly NSWCCA 16 December 1992; J (1992) 64 ACrimR 441 at 459; Veen v The Queen (No 2) (1987-88) 164 CLR 465; Hallocoglu (1991) 29 NSWLR 67.

Therefore criminal statistics on the incidence of an offence, including police data, court data and victim surveys are not “evidence” within s12 of the Criminal Appeal Act 1912 nor s29A of the Criminal Appeal Act 1986. Whilst academic literature on the effects of an offence including both physical and psychological effects, may not be “evidence” within these sections, (Hulme J dissenting), it is not material to which the Court should have regard in the individual cases as the gravity of each case is determined by, inter alia, the effects on the particular victims.

Per Hulme J:
Observations on criminal statistics.

Need for Guidelines: Armed Robbery
The increased incidence of a particular offence is a relevant factor when considering whether or not a guideline judgment is appropriate and, if so, what change with respect to past sentencing practice is called for. Whilst armed robbery is a relatively rare crime, its rate of occurrence has significantly increased in recent years.

Furthermore, the sentencing statistics of the Judicial Commission of New South Wales indicate:
(i) Non-custodial sentences are not confined to exceptional cases
(ii) Leniency is suggested in the full terms
(iii) The pattern of leniency is reinforced by the extent to which the statutory ratio between minimum and additional terms has been varied
((iv) The large proportion of lenient sentences suggest inconsistency in sentencing practices.

The sentencing statistics suggest both inconsistency and systematic leniency, which justify the promulgation of a guideline judgment.

Sentencing Guidelines: Armed Robbery
(a) Type of penalty
Armed robbery is not simply a crime against property. It is a crime against persons. The fear engendered by the perpetrator of this crime, together with the continued adverse effects on its victims, establish armed robbery to be a serious crime which required condign punishment. Non-custodial sentences for this offence should only be imposed in exceptional circumstances: Roberts (1994) 73 ACrimR 306 at 308. The number of cases in which non-custodial sentences have been imposed indicates that this principle has not been implemented by sentencing judges. The sentencing statistics suggest that this approach is also reflected in the length of custodial sentences when imposed. (Waldron NSWCCA 3 March 1994. See also Ellis (1993) 68 ACrimR 449 at 462; Lawson NSWCCA 12 December 1997).

(b) Guideline precedents
Guidelines for armed robbery sentencing in English, New Zealand, Hong Kong, Canadian, Western Australian, South Australian and Queensland authorities referred to. (Turner (1975) 61 CrAppR 67; Gould (1983) 5 CrAppR(S) 72; Attorney-General’s References Nos 3, 4, 8, 9, 10, 11 and 16 of 1990 (Dickson & Ors) (1991) 92 CrAppR 166; Attorney-General’s Reference No 14 of 1991 (Reed) (1992) 13 CrAppR(S) 446; Attorney-General’s Reference NO 7 of 1992 (Khan) (1993) CrAppR(S) 122; Moananui (1983) NZLR 537; Mo Kwong Sang (1981) HKLR 610; Brennan and Jensen (1975) 11 NSR (2d) 541; Chaisson (1975) 24 CC (2d) 159; Johnas (1982) 2 CCC (3D) 490; Norman WACCA 1 February 1989; Miles (1997) 17 WAR 518; Spiero (1979) 22 SASR 543; Dube (1987) 456 SASR 118; Fermaner (994) 61 SASR 447; Drumgoon SACCA 20 November 1995; Hammond (1996) 92 ACrimR 450).

(c) Guidelines
A category of case which is sufficiently common for the purposes of determining a guideline comprises the following elements:
(i) Young offender with no or little criminal history
(ii) Weapon like a knife, capable of killing or inflicting serious injury
(iii) Limited degree of planning
(iv) Limited, if any, actual violence but a real threat thereof
(v) Victim in a vulnerable position such as a shopkeeper or taxi driver
(vi) Small amount taken
(vii) Plea of guilty, the significance of which is limited by a strong Crown case.

Per Simpson J:
A plea of guilty should not be included in the definition of the kind of offence to which the guideline applies.

(d) The appropriate range
A sentencing range is appropriate in relation to this offence because the seven identified characteristics do not exhaust the factors relevant to sentencing, and many of the characteristics contain within themselves an inherent variability.

Sentences for an offence of this character should generally fall between four and five years for the full term. Aggravating and mitigating factors will justify a sentence below or above the range, which is itself a starting point.

Per Hulme J:
The guideline should be set so as to include a total term of penal servitude of 5 years and, except in unusual circumstances a minimum term of 3 years.

(e) Circumstantial factors
A number of circumstances are particular to the offence of armed robbery, including: (a) nature of the weapon; (b) vulnerability of the victim; (c) position on a scale of impulsiveness/planning; (d) intensity of threat, or actual use, of force; (e) number of offenders; (f) amount taken; and (g) effect on victim/s.

Per Spigelman CJ:
Drug addiction is a circumstance relevant to the sentencing exercise, but it is not itself a mitigating factor. The existence of a causal relationship between drug addiction an the commission of an offence should not automatically result in a lesser sentence. Valentini (1989) 46 ACrimR 23 at 25; Halewyn (1984) 12 ACrimR 202; Nolan [1998][ VSCA 135; Douglas (1995) 56 FCR 465 at 470; Spiero (1979) 22 SASR 543 at 549; Terizakis (1986) 41 SASR 252 at 256; Hammond (1996) 92 ACrimR 450 at 455-456, 467.

Per Wood CJ at CL:
Discussion of general principles of deterrence and drug addiction. Drug addiction is not analogous to mental abnormality in respect of which the element of general deterrence may be given less weight. The relevant principles are as follows:
(a) The need to acquire funds to support a drug habit is not an excuse to commit an armed robbery, and of itself not a matter of mitigation.
(b) The fact that an offence is motivated by such a need may be taken into account as a factor relevant to the objective criminality of the offence, insofar as it may throw light on matters such as (i) the impulsivity of the offence and the extent of any planning for it; (ii) the existence or non-existence of any alternative reason that may have operated in aggravation of the offence; and (iii) the state of mind or capacity of the offender to exercise judgment.
(c) It may also be relevant as a subjective circumstance, insofar as the origin or extent of the addiction, and any attempts to overcome it, might (i) impact upon the prospects os recidivism/rehabilitation; (ii) suggest that the addiction was not a matter of personal choice but was attributable to some other event for which the offender was not primarily responsible; and (iii) justify special consideration in the case of offenders to be at the “cross roads”: Osenkowski (1982) 5 ACrimR 394.

Per Simpson J:
Discussion of general principles of rehabilitation and drug addiction. Where a combination of the following two circumstances exist, then general deterrence and other sentencing objectives such as retribution, may yield to rehabilitation: (i) a background to the addiction that may explain the offender’s decision to use drugs, and diminish his or her moral culpability for that decision; (ii) demonstrated progress towards rehabilitation.
- 139 -
IN THE COURT OF
CRIMINAL APPEAL


60559/98
60558/98
60561/98
60746/98
60596/98
60595/98
60511/98


                                SPIGELMAN CJ
                                WOOD CJ at CL
                                NEWMAN J
                                HULME J
                                SIMPSON J


                                Wednesday 12 May 1999

                REGINA v Paul Anthony HENRY
                REGINA v Stephen Anthony BARBER
                REGINA v Hoai Vinh TRAN
                REGINA v Troy David SILVER
                REGINA v Theo TSOUKATOS
                REGINA v Bill KYROGLOU
                REGINA v John David JENKINS

JUDGMENT

1 SPIGELMAN CJ: In Jurisic NSWCCA 12 October 1998, this Court indicated that it would in future be prepared to issue guideline judgments with respect to sentencing for particular offences. The Court has listed and heard together six Crown appeals and one severity appeal, six of which involves the offence of armed robbery under s97(1) of the Crimes Act 1900 (NSW) and one, the aggravated offence under s97(2). The Crown has submitted that it is appropriate to promulgate a guideline judgment with respect to this offence.
2 In Jurisic, in a judgment with which the other four members of the Court agreed, I said:
“In my opinion, guideline judgments should now be recognised in New South Wales as having a useful role to play in ensuring that an appropriate balance exists between the broad discretion that must be retained to ensure that justice is done in each individual case, on the one hand, and the desirability of consistency in sentencing and the maintenance of public confidence in sentences actually imposed, and in the judiciary as a whole, on the other.” (14)
        “Such guidelines are intended to be indicative only. They are not intended to be applied to every case as if they were rules binding on sentencing judges. Decisions of appellate courts on sentencing are not to be treated as binding precedents.” (14-15)
3 In Jurisic I referred to the numerous cases in which this Court has made statements of general principle with a view to guiding sentencing practice in trial courts. The promulgation of formal guideline judgments, labelled as such, was a development of that practice.
4 In Jurisic, Wood CJ at CL said:
“The Court has … over the years endeavoured to lay down sentencing principles for particular classes of case where sentences reflecting a significant element of general deterrence are required, or when non-custodial options are inappropriate. It appears that sometimes these principles are lost or that their significance is overlooked, in the volume of appellant decisions handed down and in the pressures imposed on trial courts to dispose of increasingly busy criminal lists.
        By tagging selected decisions as guideline judgments, the Court is not to be taken as usurping the function of the legislature, or as inappropriately intruding into the exercise of the sentencing discretion reserved to trial judges. Rather, what is intended is for the Court of Criminal Appeal to highlight the sentencing principles which fall for it to determine, in a way that might assist trial judges, the DPP and trial counsel, and reduce the occasion for that degree of inconsistency or departure from principle that is an indicator of injustice.” (1-2)
5 Subject to any relevant statutory requirements, the sentencing task involves the exercise of a broad discretion which, centuries of practical experience strongly indicate, is best conferred on trial judges.
6 The circumstances in which it is appropriate for appellate courts to interfere with discretionary decisions of this character are confined. In the case of Crown appeals against sentence, even more stringent restrictions have been applied.
7 The ineluctable core of the sentencing task is the process of balancing overlapping and contradictory objectives. At the appellate level, that characteristic extends to the balancing of the objectives of consistency and individualisation. The Court must sentence both the offender and the offence.
8 During the course of these proceedings the Respondents to the Crown appeals and the Appellant in the severity appeal, relied on certain observations of Mahoney ACJ in Lattouf (NSWCCA 12 December 1996) where his Honour repeated his own remarks in Kable v DPP (1995) 36 NSWLR 374 at 394:
“If justice is not individual, it is nothing.”
9 This ringing phrase must not be taken out of context. In Lattouf his Honour emphasised the multiple objectives served by the sentencing process. One could equally well say “If justice is not consistent, it is nothing”.
10 As His Honour put it in Lattouf:
“General sentencing principles must be established, so that the community may know the sentences which will be imposed and so that sentencing judges will know the kind and the order of sentence which it is appropriate that they impose. But, of course, principles are necessarily framed in general terms. General principles must, of their nature, be adjusted to the individual case if justice is to be achieved. For this reason, it is in my opinion important in the public interest that the sentencing process recognise and maintain a residual discretion in the sentencing judge … There is a public interest in the adoption and articulation of sentencing principles which will deter the commission of serious crime and punish those who commit it … But there are other interests to which the sentencing process must have regard; these are other objectives which the sentencing process must seek to achieve. Paramount amongst these is the achievement of justice in the individual case.” (7)
11 There is no conflict between the system of guideline judgments established by Jurisic and the reasoning of Mahoney ACJ in Lattouf.


Guidelines: General
12 As I indicated in Jurisic, the purpose of a guideline judgment is to foster consistency in sentencing. The importance of consistency was well expressed by Mason J in Lowe (1994) 154 CLR 606:
“Just as consistency in punishment - a reflection of the notion of equal justice - is a fundamental element in any rational and fair system of criminal justice, so inconsistency in punishment, because it is regarded as a badge of unfairness and unequal treatment under the law, is calculated to lead to an erosion of public confidence in the integrity of the administration of justice.” (610-611)
13 The appropriateness of an appellate court establishing guidelines has been authoritatively established. In Norbis v Norbis (1986) 161 CLR 513, the High Court had before it an issue concerning the power of the Full Court of the Family Court to lay down guidelines with respect to the exercise of statutory discretions by trial judges. By majority, the Court held that the promulgation of such guidelines was permissible.
14 Mason and Deane JJ said:
“It has sometimes been said by judges of high authority that a broad discretion left largely unfettered by Parliament cannot be fettered by the judicial enunciation of guidance in the form of binding rules governing the manner in which the discretion is to be exercised. However, it does not follow that, because a discretion is expressed in general terms, Parliament intended that the Court should refrain from developing rules or guidelines affecting its exercise. One very significant strand in the development of the law has been the judicial transformation of discretionary remedies into remedies which are granted or refused according to well settled principle. It has been a development which has promoted consistency in decision making and diminished the risks of arbitrary and capricious adjudication. The proposition referred to in the beginning of this paragraph should not be seen as inhibiting an appellate court from giving guidance, which falls short of constituting a binding rule, as to the manner in which the discretion should be exercised. And despite the generality of some of the statements to which we have referred, there may well be situations in which an appellate court will be justified in giving such guidance the force of a binding rule by treating a failure to observe it as constituting grounds for a finding that the discretion has miscarried.
        The point of preserving the width of the discretion which Parliament has created is that it maximises the possibility of doing justice in every case. But the need for consistency in judicial adjudication, which is the antithesis of arbitrary and capricious decision making, provides an important countervailing consideration supporting the giving of guidance by appellate courts, whether in the form of principles or guidelines. The tension between the two considerations, each of fundamental importance in family law, has inevitably led to a near dilemma for the Full Court of the Family Court. To avoid the risk of inconsistency and arbitrariness which is inherent in the system of relief involving a complex of discretionary assessments and judgments, the Full Court, as a specialist appellate court with unique experience in family law in this country, should give guidance as to the manner in which these assessments and judgments are to be made. Yet guidance must be given in a way that preserves, so far as it is possible to do so, the capacity of the Family Court to do justice according to the needs of the individual case, whatever its complications may be. Reconciliation of these goals suggests that in most, if not all, cases the Full Court of the Family Court should give guidance in the form of guidelines rather than binding principles of law.” (519-520 internal references omitted)
15 The other member of the majority in Norbis was Brennan J. He agreed with the reasons of Mason and Deane JJ with one exception:
“The proposition with which I am unable to agree is this: that an appellate court that gives guidance as to the manner in which a statutory discretion should be exercised may prescribe that such guidance should have the force of a binding legal rule. The authority of an appellate court to give guidance is not to be doubted. It is inevitable that the wisdom gained in continually supervising the exercise of a statutory discretion will find expression in judicial guidelines. That is not to invest an appellate court with legislative power but rather to acknowledge that, in the way of the common law, a principle which can be seen to be common to a particular class of case will ultimately find judicial expression. The orderly administration of justice requires that decisions should be consistent one with another and decision making should not be open to the reproach that it is adventitious … An unfettered discretion is a versatile means of doing justice in particular cases, but unevenness in its exercise diminishes confidence in the legal process.” (536)
16 His Honour went on to say:
“It is one thing to say that principles may be expressed to guide the exercise of a discretion; it is another thing to say that the principles may harden into legal rules which would confine the discretion more narrowly than the Parliament intended.” (537)
17 His Honour also said:
“There may well be situations in which an appellate court will be justified in setting aside a discretionary order if the primary judge, without sufficient grounds, has failed to apply a guideline in the particular case. Where there is nothing to mark the instant case as different from the generality of cases, the failure will suggest that the discretion has not been soundly exercised. The distinction between such a guideline and a binding rule of law, though essential, may be thin in practice. But the distinction must be maintained and a failure to apply the guideline cannot be treated as an error of law: a failure to apply the guideline is no more than a factor which warrants a close scrutiny to the particular exercise of the discretion. What cannot be shut out is the discretion of a primary judge not to apply the guideline when the circumstances of the particular case show that its application would produce an unjust or inequitable result or that another approach would produce a more just and equitable result.
        The only compromise between idiosyncrasy in the exercise of the discretion and an impermissible limitation of the scope of the discretion is to be found in the development of guidelines from which a judge may depart when it is just and equitable to do so - guidelines which are not rules of universal application, but which are generally productive of just and equitable orders. If it is possible to develop such guidelines, it is possible to ensure order and consistency in the exercise of the discretionary jurisdiction under the Family Law Act.” (537-538)
18 The appropriateness of an appellate court laying down guidelines for the exercise of a statutory discretion has also been noted in a range of cases, invariably with reference to the desirability of consistency in judicial decision making. These cases include:
· the statutory discretion to award costs in criminal proceedings (Latoudis v Casey (1990) 170 CLR 534 at 541-542, 558-559, 562).
· the exercise of the discretion to award costs in civil matters (Oshlack v Richmond River Council [1998] HCA 11; 72 ALJR 578; 152 ALR 83 at [35], [134] and [65])
· the exercise of a discretion to order the winding up of a company (FAI Insurances Limited v Goldleaf Interior Decorators Pty Ltd (No 2) (1988) 14 NSWLR 644 at 646-647, 657-658 and 660-661).
· the exercise by a licensing magistrate of a discretion to grant a liquor licence (Shreeve v Martin (1969) 72 SR(NSW) 279 at 289-291).
· the exercise of a statutory discretion to grant leave to commence proceedings out of time (Salido v Nominal Defendant (1993) 32 NSWLR 524 at 532-533, 535-539, 541).
· The exercise of a discretion to grant leave to appeal from decisions of arbitrators (Pioneer Shipping Limited v BTP Tioxide Limited [1982] AC 724 at 742-743; FAI Insurances v Goldleaf Interior Decorators supra at 661; Leighton Contractors Pty Limited v Kilpatrick Green Pty Limited [1992] 2 VR 505 at 516, 517, 521).
19 The High Court has confirmed that a Court may establish guidelines for the exercise of an inherent jurisdiction, namely the parens patriae jurisdiction to act in the best interest of the child. (Secretary, Department of Health and Community Services v JWB and SMB (1991) 175 CLR 218 at 260).
20 The Full Court of the Supreme Court of Victoria has applied the approach to guidelines in Norbis to the determination of guiding principles for the exercise of a discretionary power of the Court in its inherent jurisdiction, namely the power to order the dismissal of proceedings for want of prosecution. (Maysell v Transport Industries Insurance Co Limited [1995] 2 VR 328 at 334-335).
21 In the case of matters covered by common law principles - like sentencing for New South Wales offences - there is no inhibition which may arise from the prospect of confining a statutory discretion, which the Parliament has conferred in an unconfined form. (Special considerations will arise in formulating guidelines for Commonwealth offences by reason of s16A of the Crimes Act 1914).
22 As noted above, in Norbis the joint judgment of Mason and Deane JJ contemplated the possibility of a guideline having “the force of a binding rule”, with the consequence that failure by a trial judge to apply a guideline may constitute a ground for finding that the exercise of the discretion miscarried. It was in this respect that Brennan J disagreed.
23 The Full Court of the Family Court has adopted the option contemplated by Mason and Deane JJ:
“In our view the time has come to regard a departure from a longstanding guideline … without adequate explanation as a ground for finding that the exercise of discretion has miscarried.” Docters Van Leeuwen (1990) 14 FamLR 130 at 134.
        (See also Lalor (1989) 14 FamLR 282 at 285; Joshua (1997) 22 FamLR 203 at 214; Nygh “Should the Full Court Offer More Guidance to Judges Sitting at First Instance” (1993) 7 Australian Journal of Family Law 137 esp at 146-149).
24 However, a differently constituted Full Court of the Family Court has stated that guidelines developed by the Full Court should not be regarded as rules of law:
“The guidelines that we propose are simply guidelines; they are not rigid rules of law and it does not follow that a departure from them will necessarily vitiate a judgment, although judges, judicial registrars and registrars should, we think, give sufficient reasons for departing when they consider a departure is appropriate.” (Re K (1993) 117 FLR 63 at 81).
25 The balance of authority strongly supports the reasoning of Brennan J in Norbis. See Shreeve v Martin supra at 290 per Walsh JA; FAI Insurances v Goldleaf Interiors supra at 661 per McHugh JA; Latoudis v Casey supra at 558-559 per Dawson J, with whom Brennan J agreed; Leighton Contractors v Kilpatrick Green supra at 516-517 per Fullagar J with whom McGarvie and Ashbee JJ agreed; Maysell v Transport Industries Insurance supra at 334-335 per Brooking, Teague and Hedigan JJ; Oshlack v Richmond River Council supra at [35] per Gaudron and Gummow JJ, and at [134] per Kirby J.
26 A similar issue has arisen in Canada. The Supreme Court of Canada has unanimously affirmed that it is appropriate for a criminal appellate court to lay down guidelines in the nature of a starting point for sentencing for a particular offence. McDonnell (1997) 114 CCC (3d) 436 at pars 43, 58-61. (See generally Bloos and Renke “Case Comment: Stopping Starting Points R v McDonnell” (1997) 35 Alberta LR 795).
27 In McDonnell both (majority and minority) judgments rejected the proposition that departure from a previously decided starting point could, itself, constitute an error of principle that would justify appellate intervention ( at pars 32-34; 100, 109).
28 The majority accepted that departure from a starting point may “suggest” that a sentence is outside the permissible range but is not “determinative” of that proposition (par 43). The minority seemed to regard departure from the starting point as being more strongly indicative of legal error, in terms of manifest inadequacy (or “demonstrably unfit” in the Canadian terminology) (pars 98-101). Neither judgment adopted the approach which Mason and Deane JJ contemplate in Norbis.
29 A guideline judgment on the subject of sentencing should not lay down a requirement or anything in the nature of a rule. The failure to sentence in accordance with a guideline is not itself a ground of appeal. Guidelines are not rules of universal application. They may be departed from when the justice of a particular case requires such departure. I made this clear in Jurisic when I said:
“Such guidelines are intended to be indicative only. They are not intended to be applied in every case as if they were rules binding on sentencing judges. Decisions of appellate courts on sentencing are not to be treated as binding precedents.” (15)
30 In this respect I refer to my adoption in Jurisic (at 15) of the analysis by Dunn LJ in De Havilland (1993) 5 CrAppR(S) 109 at 114, to the effect that decisions on sentencing are not authorities binding on lower courts in the way decisions on substantive law are binding. I went on to say:
“ … such guidelines are not binding in any formal sense. They represent a relevant indicator, much as trial judges have always regarded statutory maximum penalties as an indicator”. (Jurisic at 16)
31 Nevertheless, where a guideline is not to be applied by a trial judge, this Court would expect that the reasons for that decision be articulated, so that the public interest in the perception of consistency in sentencing decisions can be served and this Court can be properly informed in the exercise of its appellate jurisdiction.
Guidelines: Sentencing
32 In Jurisic I outlined the development of sentencing guideline judgments in England. It is not necessary to repeat that analysis.
33 In Canada, a starting point approach to sentencing guidelines has been endorsed by the Supreme Court in McDonnell supra. McLachlin J said:
“The starting point approach to sentencing involves two steps. First, the judge determines the range of sentences for a typical case. Using that range as a starting point, a trial judge then adjusts the sentence upward or downward on the basis of factors relating to the particular offence and offender … the starting point approach combines general considerations relating to the crime committed with personalised considerations relating to the particular offender and the unique circumstances of the assault.” (par 58)
34 Her Ladyship went on to say:
“The traditional notion that sentencing is primarily a matter of impression for the sentencing judge and only secondarily a matter of principle began to be questioned by the Courts in the mid 60s. Behind the challenge lay increasing recognition that some measure of uniformity was essential in a sentencing process that was not only just, but was perceived to be just.” (par 65)
        and
“The starting point approach appears to meet both the requirements of uniformity and consistency in sentencing and individualised justice.” (par 78)
        (See also pars [23-24, 43]; Ruby Sentencing (4th ed, 1994) esp pp481-482).
35 The judgments in the Supreme Court referred with approval to the decision of the Alberta Court of Appeal in Sandercock (1985) 22 CCC (3d) 79 where the Court said:
“The sentencing process now adopted by this court is to state typical categories with precision, and to acknowledge at the same time that each actual case presents differences from the archetypical case. These differences might mitigate or aggravate.” (83)
36 This is similar to the approach adopted by this Court in Jurisic.
37 The practice of the Full Court of the Supreme Court of South Australia is to promulgate sentencing standards in the form of an “appropriate sentence range”. However, departure from the range is not itself error.
38 In Police v Cadd (1997) 94 ACrimR 466, Doyle CJ said:
“It is the function of this Court to ensure that the sentencing process in individual cases is not affected by error of principle. It is also the function of this Court to ensure that sentences are neither excessive nor inadequate. The latter function is performed in two ways. First, in individual cases, by correcting a particular sentence that is considered to be excessive or inadequate. Secondly, by establishing standards of sentencing for particular offences, when the court thinks it appropriate to do so. This may be done over time through the process of correcting individual sentences. But it may also, in my opinion, be done by the court indicating an appropriate sentence range for a particular offence or offences of a particular type.” (479-480)
39 His Honour went on to cite a number of cases in which the Court had indicated appropriate sentencing ranges and referred, with approval to the following remarks of Cox J in King (1988) 48 SASR 555 at 557-558; 34 ACrimR 412 at 414-415:
“… this case is about sentencing standards, but it is important, I think, to bear in mind that when a sentence is created, either by cumulative force of individual sentences or by a deliberate act of policy on the part of the Full Court, there is nothing rigid about it. Such standards are general guides to those who have to sentence in the future …”
40 Doyle CJ went on to say in Cadd:
“The function of establishing appropriate standards is an important aspect of ensuring, as best one can, that adequate standards of punishment are observed. Establishing appropriate standards also tends to ensure that there is such consistency of approach as is achievable in a system in which the appropriate sentence depends upon, in part, the circumstances of the individual case and of the individual offender, and a system in which sentencing is as individualised as it is in our system.” (480)
        (See also 487 per Duggan J; 488 and 490-491 per Mullighan J; 511 per Lander J; 520 per Bleby J).
41 I express my agreement with the following observations of Lander J:
“There are a number of reasons why this Court can and should indicate in a general way that a particular class of offences should be dealt with within certain standards. First it gives general guidance to judges of this Court and to the courts below of what the superior court within the State considers an appropriate range of penalties for a particular class of offence. Secondly an indication by this Court of an appropriate standards is likely to best ensure consistency of penalty throughout the State. Thirdly it is, in my opinion, part of the deterrent process that this Court pronounces, not only for the assistance of the courts below, but for the general education of the public and the particular education of those who may be likely to offend. The public should know in advance that offences of a particular kind will be likely to lead to a penalty in a range indicated by this Court.
        I agree therefore that it is appropriate for this Court to indicate a standard in relation to some offences. However not all offences are capable of being categorised in that manner. Some offences such as assaults and manslaughter depend so very much upon the circumstances of the particular offence that they cannot be considered in a general way as being within any particular sentence range.” (511)
42 The South Australian practice was affirmed by the High Court when rejecting a special leave application from one of four sentencing standard cases which had been heard together and which I will consider further below. Brennan J said:
“By their decisions, the Court of Criminal Appeal hoped to ‘remind sentencing judges of the great importance of maintaining adequate standards of punishment in sentencing for armed robbery’. That being the object of the Court of Criminal Appeal, this Court should not grant special leave to review the range of sentences which the Court of Criminal Appeal has set.
        However, the Court of Criminal Appeal is bound to apply general principles of sentencing to any case in which the Crown seeks to have a range of sentencing established or confirmed so that the actual sentence in any case properly reflects its unique circumstances.” (Bini (1994) 68 ALJR 859)
        Statutory Scheme for Robbery
43 Chapter 1 of Pt 4 of the Crimes Act 1900 (NSW) creates a hierarchy of offences with respect to robbery, an offence distinguished from other forms of stealing by an element of threat or force, putting a person in fear:
Maximum Penalty 14 years
        (i) Robbery or assault with intent to rob from the person of another (s94).
        Maximum Penalty 20 years
        (ii) The offence in (i), in circumstances of aggravation, which includes use of violence or malicious infliction of actual bodily harm or deprivation of liberty (s95).
        (iii) Robbery or assault with intent to rob being armed with an offensive weapon or in company (s97(1)).
        Maximum Penalty 25 years
        (iv) The offence in (ii), with actual wounding or infliction of bodily harm (s96).
        (v) The offence in (iii), when armed with a dangerous weapon (s97(2)).
        (vi) Wounding or inflicting grievous bodily harm before, at the time of or immediately after a robbery, when armed with an offensive weapon (s98).
44 In Brown (1989) 17 NSWLR 472 in a joint judgment of Gleeson CJ, Newman and Loveday JJ, this Court said:
“As is the case in relation to various other subject matters, the provisions of the Crimes Act 1900 dealing with robbery establishes a series of offences, in ascending degrees of seriousness, and with ascending orders of maximum penalty, depending upon the circumstances of the case … It is apparent that within that range of offences there may exist an enormous variety of individual facts and circumstances which will accompany the commission of a particular crime.” (473)
45 The cases now before the Court are concerned only with s97(1), save for one conviction under s97(2). The section provides:
“97(1) Whosoever, being armed with an offensive weapon, or instrument, or being in company with another person, robs or assaults with intent to rob, any person, or stops any mail, or vehicle, railway train, or person conveying a mail, with intent to rob, or search the same, shall be liable to penal servitude for 20 years.
        (2) A person is guilty of an offence under this subsection if the person commits an offence under subsection (1) when armed with a dangerous weapon. A person convicted of an offence under this subsection is liable to penal servitude for 25 years.”
46 Section 4(1) defines “offensive weapon or instrument” to:
“… include a dangerous weapon and also include an imitation or replica of an offensive weapon or of an offensive weapon, or an instrument as the case may require.”
47 “Dangerous weapon” means a firearm, a prohibited weapon within the Prohibited Weapons Act 1989 (NSW) or a spear gun.
48 It is with respect to the offence under s97(1) that the Court was invited by the Crown to promulgate a guideline judgment.
Crown Material
49 The Crown sought to put before the Court a range of materials which were not before the sentencing judge in any of the cases. It is material of a general character relevant to sentencing which can be, and sometimes is, put before a sentencing judge. This material was said by the Crown to be relevant to the determination of the individual cases, as well as to the formulation of sentencing guidelines of general application.
50 The Crown also put before the Court sentencing statistics for s97(1) as compiled by the Judicial Commission of New South Wales. No objection was taken to the Court referring to this material, either in the individual cases or for purposes of the formulation of guidelines. Such information is, of course, available to all sentencing judges and, accordingly, this Court can act on the basis that the trial judge either had, or could have had, access to it if he or she wished. The Court of Criminal Appeal must, of course, be in the same position in this regard. All parties conceded that access to this information was appropriate.
51 The Respondents to the Crown appeals and the Appellant in the severity appeal objected to the Court taking into account the additional material in the individual cases. This material was of three kinds:
        (i) Crime statistics on the incidence of armed robbery offences, including police data, court data and victim surveys.
        (ii) Academic literature on the effects of armed robbery on victims, including both physical and psychological effects.
        (iii) A report by a clinical psychologist, prepared for purposes of this hearing, on the psychological impact of armed robbery upon victims.
52 The objection to the use of this evidence for the individual cases was based on the proviso to s12(1) of the Criminal Appeal Act 1912. That section is in the following terms:
“(1) The court may, if it thinks it necessary or expedient in the interests of justice:
        (a) order the production of any document, exhibit, or other thing connected with the proceedings, and
            (b) order any persons who would have been compellable witnesses at the trial to attend and be examined before the court, whether they were or were not called at the trial, or order any such persons to be examined before any judge of the court or before any officer of the court, or justice, or other person appointed by the court for the purpose, and admit any deposition so taken as evidence, and
            (c) receive the evidence, if tendered, of any witness (including the appellant) who is a competent, but not a compellable witness, and
            (d) where any question arising on the appeal involves prolonged examination of documents or accounts, or any scientific or local investigation, which cannot, in the opinion of the court, be conveniently conducted before the court, the court or any judge thereof may refer the question for inquiry and report to a commissioner appointed by the court, and act upon the report of any such commissioner so far as the court thinks fit, and
            (e) appoint any person with special expert knowledge to act as assessor to the court in any case in which it appears to the court that such special knowledge is required for the determination of the case;
            and exercise in relation to the proceedings of the court any other powers which may for the time being be exercised by the Supreme Court on appeals or applications in civil matters, and issue any warrant or other process necessary for enforcing the orders or sentences of the court: Provided that in no case shall any sentence be increased by reason of, or in consideration of any evidence that was not given at the trial. (Emphasis added)
53 Section 12 was part of the original Criminal Appeal Act when it was enacted in 1912. The Act as at that time did not make any provision for appeals by the Crown. By s6(3) of the Act as originally enacted, this Court was expressly empowered to increase the sentence on an appeal by a convicted person. This necessitated the proviso in s12.
54 This is the explanation for the proviso, with which I agree, given by McPherson J in Beldan (1986) 21 ACrimR 159 at 167, where the Court was concerned with the equivalent provision in s671B of the Criminal Code (Qld), inserted in 1913. I also agree with his Honour’s conclusion that the proviso to s12(1) applies to Crown appeals, even though they were subsequently introduced.
55 In Behar NSWCCA 14 October 1998, I made reference to the possible difficulty which s12 may present for the preparation of guideline judgments. Behar was handed down a few days after Jurisic. Subsequently the Parliament amended the Criminal Procedure Act 1986 with the introduction of Pt 8 “Sentencing Guidelines” by the Criminal Procedure Amendment (Sentencing Guidelines) Act 1998. Section 26 of the Act now empowers the Attorney General to apply to the Court of Criminal Appeal for a guideline judgment.
56 Section 29A now provides:
“(1) Nothing in section 12 of the Criminal Appeal Act 1912 limits the evidence or other matters that the Court may take into consideration in giving a guideline judgment (whether or not on application under section 26) and the Court may inform itself as it sees fit.
        (2) The Court must not increase a sentence in any appeal by reason of, or in consideration of, any evidence that is used by the Court in giving a guideline judgment in the appeal and that was not given at the trial.”
57 By reason of the words “whether or not on application under s26” in 29A(1), the material on which the Crown sought to rely, can be taken into account by this Court in preparing a guideline judgment with effect in futuro. This was conceded by all parties.
58 Section 29A puts beyond doubt, in the context of the promulgation of sentencing guidelines, what would probably have been accepted even without legislative provision. The formulation of guidelines, even with respect to the exercise of statutory discretions, often requires information which goes beyond the circumstances of an individual case.
59 The recognition that judges do in fact develop and change the law, as distinct from merely declaring the law, has received general acceptance comparatively recently. Accordingly, the circumstances in which the Court will receive material, including evidence in the formal sense, for purposes of such policy development have never been clearly articulated and established. (See Davis “The Judiciary - Maintaining the Balance” in Finn (ed) Essays on Law and Government vol 1 pp283-285; Doyle “Implications of Judicial Law Making” in Saunders (ed) Courts of Final Jurisdiction: The Mason Court in Australia esp at pp96-97). It is plain that cases which raise policy issues cannot be allowed to become law reform commission inquiries.
60 The means of acquiring information for the purposes of policy development should not be confined by the rules of evidence developed for fact finding with respect to matters that only concern the parties to a particular case. (See Davis “An Approach to Problems of Evidence in the Administrative Process” (1942) 55 HarvLRev 364 esp at 402; Davis “Judicial Notice” (1955) 55 Col. LR 945 esp at 952-953; Carter “Judicial Notice” in Campbell and Woller (eds) Well and Truly Tried (1982) at pp92-94; Ligertwood Australian Evidence (3rd ed, 1998) par 6.42).
61 Similar issues have arisen with respect to the determination of constitutional facts. As Brennan J has put it:
“There is a distinction between a judicial finding of a fact in issue between parties upon which a law operates to establish or deny a right or liability and a judicial determination of the validity or scope of a law when its validity or scope turns on a matter of fact. When a court, in ascertaining the validity or scope of a law, considers matters of fact, it is not bound to reach its decision in the same way as it does when it tries an issue of fact between a parties. The validity and scope of a law cannot be made to depend on the course of private litigation.” (Gerhardy v Brown (1984) 159 CLR at 141-142).
62 His Honour went on to refer to Breen v Sneddon (1961) 106 CLR at 406-406 and Commonwealth Freighters Pty Ltd & Boland v Sneddon (1959) 102 CLR 280 at 292 and added:
“The court may, of course, invite and receive assistance from the parties to ascertain the statutory facts, but it is free also to inform itself from other sources. Perhaps those sources should be public or authoritative, and perhaps the parties should be at liberty to supplement or controvert any factual material on which the court may propose to rely, but these matters of procedure can wait consideration on another day. The court must ascertain the statutory facts ‘as best it can’ and it is difficult and undesirable to impose an a priori restraint on the performance of that duty.”
63 This reasoning is applicable to the determination of sentencing principles or the promulgation of a guideline judgment operative in futuro.
64 Different considerations apply in the sentencing context, when material relevant at a policy level is sought to be relied upon, either directly or indirectly, by the application of the new or differently articulated policy, principle or guideline, to the case on appeal. Issues of double jeopardy and disparity arise in this context.
65 The Respondents to the Crown appeal, and the Appellant in the severity appeal, rely on the proviso to s12(1) of the Criminal Appeal Act, as reinforced by s29A(2) of the Criminal Procedure Act 1986. These provisions reflect the principle of double jeopardy. The Respondents submit that none of the material should be taken into account with respect to the particular cases before the Court.
66 No submission was made that either the criminal statistics or the academic literature could not have been taken into account by the sentencing judge. The focus of the submissions was on the double jeopardy element, reflected in the two statutory provisions. Accordingly, it is unnecessary to consider whether Re Richardson [1920] SALR 24, which turns on the “generally known” formulation from Holland v Jones (1917) 23 CLR 149, is still good law in the light of the different formulation in s144 of the Evidence Act 1995.
67 Nor was any submission made that, in any case, there was an inadequate opportunity to deal with the additional material (cf s144(4) Evidence Act 1995).
68 The Crown eventually conceded that the report of the clinical psychologist expressly prepared for use in these proceedings, cannot be relied on in the individual cases. This leaves two categories of material:
        (i) Crime statistics
        (ii) Academic literature on victim impact.
69 The issue turns on the proper construction of the word “evidence” in each of the proviso to s12(1) of the Criminal Appeal Act and s29A(2) of the Criminal Procedure Act. It is plain that in the later Act, the Parliament intended to use the word in the same sense as it was used in the proviso to the former Act.
70 It is noteworthy that there is a distinction between subs (1) and subs (2) of s29A of the Criminal Procedure Act in this regard. Subsection (1) refers both to “evidence” and “other matters”, whereas subs (2) refers only to “evidence”.
71 The principal thrust of the submissions for the Respondents to the Crown appeals was that the additional material on which the Crown wished to rely is “evidence” because the Crown seeks to prove as facts the following:
        (i) The existence of an increase in the number of armed robberies in recent times.
        (ii) The extent and intensity of the impact of armed robberies on victims.
72 Both of these, it is said, are issues of “fact” about which the Crown seeks to adduce “evidence” in the form identified.
73 Merely to identify an issue as one of “fact” does not lead to the conclusion that the Court is restricted to “evidence” with respect to its determination. For example, the meaning of an ordinary English word has long been held to be a matter of fact not law, yet the Court consults dictionaries and other books in order to determine the meaning. (See Australian Gaslight Company v The Valuer General (1940) 40 SR (NSW) 126 at 137).
74 An appellate court may take judicial notice on appeal of something which has not been “noticed” at first instance. (See Warren v Pilkington (1960) Tas SR 6 at 9-15). This must, of course, be subject to any contrary statutory provisions.
75 The Court may “notice” certain notorious facts of which ordinary persons are presumed to be aware. The Court can “notice” such facts either directly, or after being “reminded” of such facts by appropriate information or evidence. (See eg Holland v Jones (1917) 23 CLR 149 at 153). The Court may make reference to works of reference or authority in order to “remind it” of what it “knows”. (See McQuaker v Goddard (1941) KB 687 at 700-701). Sometimes evidence will be required.
76 The information which the Court acquires in this way is not “evidence strictly so called”. (See McQuaker v Goddard supra at 700; Baldwin & Francis Limited v Patents Appeal Tribunal [1959] AC 663 at 691; Saul v Menon (1982) NSWLR 314 at 325; Wigmore on Evidence (3rd ed) par 2568a; Stone & Wells Evidence: Its History and Policies (1991) p174; Noakes “The Limits of Judicial Notice” (1958) 74 LQR 59 at 63). Nor, in my opinion, is “knowledge”, of which proof is not required under s144 of the Evidence Act, “evidence strictly so called”.
77 The proviso to s12, and its interstate equivalents have been the subject of judicial consideration. In Beldan supra, the Queensland Supreme Court held that the proviso did not preclude the receipt by the appellate court of evidence to show that the sentencing process was affected by deception on behalf of the Respondents to a Crown appeal (Beldan supra at 167 and 178). This Court has applied Beldan and admitted evidence to show that the sentencing judge was deceived by evidence of a promise to give assistance to the Crown which, this Court found, the Respondent had no intention to honour. (See Chanh Nghia Ly NSWCCA 16 December 1992 at 10-11. See also Todhunter v Attorney-General (Com) (1994) 52 FCR 228 at 246).
78 The issue also arose in J (1992) 64 ACrimR 441, where Olsson J, with whom King CJ and Mulligan J agreed, held that the South Australian equivalent section (s359 of the Criminal Law Consolidation Act 1925 (SA)) did not prevent the receipt of evidence as to the conduct of the Respondent to a Crown appeal in failing to honour an undertaking to provide assistance to the Crown in other proceedings. His Honour said:
“There is no doubt that the clear intention of the Parliament was to ensure that, so far as possible, an offender was not unfairly placed in double jeopardy; and that, where further evidence not given at trial is admitted on appeal, that ought not, in the normal course, lead to an increase in the sentence imposed.

        When the proviso to s359 speaks of evidence that was not given at the trial it is, in my opinion, directing its attention to matters going to either the commission of the offence or the personal circumstances of the accused. It is not concerning itself with the situation in which the Crown is merely informing the court that the person sentenced has failed to honour a specific undertaking given to the court as an express basis upon which the sentence is imposed.” (459)
79 In Veen v The Queen (No 2) (1987-88) 164 CLR 465, one of the issues before the Court was whether the intermediate Court of Appeal should have received material about the state of psychiatric services in New South Wales gaols as facts relevant to the determination of sentencing principles. In the event the Court held that it was unnecessary to do so but added, by way of obiter dictum:
“It is unnecessary to consider the procedure adopted by the Court of Criminal Appeal in admitting evidence as to the state of gaol psychiatric services but, subject to appropriate procedural safeguards, we would not regard that court as being precluded from receiving further material relevant to the adoption of sentencing principles or sentencing policy of general application.” (473)
80 Notwithstanding the reference to “policy of general application”, the issue in Veen (No 2) concerned the use of any such “policy” in the particular case before the court.
81 This passage from Veen (No 2) was applied by this Court in Hallocoglu (1991) 29 NSWLR 67, where the Court received a document, over objection, being the booklet supplied to prisoners serving periodic detention outlining the operations of that scheme. The Court expressly rejected the submission that such material constituted “evidence” within the meaning of s12(1) of the Criminal Appeal Act. Hunt CJ at CL said:
“That submission must be rejected … We are entitled to inform ourselves from such material as to the true consequences of an order that a sentence be served by way of periodic detention.” (73-74)
82 It may be that when Hunt CJ at CL held in Hallocoglu that s12 did not prevent the Court informing itself about the administration of the periodic detention system, he had in mind a test for “evidence” similar to that propounded by Olsson J in J: “matters going to the commission of the offence or the personal circumstances of the accused”. Whilst Hallocoglu was, in certain respects, a guideline judgment, the use of this material was not restricted to the formulation of sentencing principles. The Court had reference to the material for the purposes of the individual case.
83 I am unable to distinguish the criminal statistics on which the Crown seeks to rely in this case, from the material relied upon by the Court in Hallocoglu.
84 In my opinion, it should now be accepted that “evidence” in s12 of the Criminal Appeal Act and s29A of the Criminal Procedure Act is confined to “matters going to the commission of the offence or the personal circumstances of the accused”. On this basis the criminal statistics are not “evidence” within the proviso.
85 The academic literature on the effects on victims is in a different position. Each individual case had its own victim or victims. The objective gravity of each case is determined by, inter alia, the effects on victims in the particular case. The general literature is only of assistance at the level of sentencing principle. Whilst it may not be “evidence” within the proviso to s12, it is not material to which the Court should have regard in the individual cases.
Crime Statistics and Need for Guidelines
86 The Crown sought to rely on crime statistics for the purpose of indicating that sentencing guidelines were required by reason of the prevalence of the offence. Prevalence is a relevant consideration when deciding an appropriate level of sentence. (See eg Giles v Barnes (1969) SASR 174 at 181; Peterson (1983) 11 ACrimR 164 at 167-168, 169; Henderson (Jana Brian) (1991) 58 ACrimR 369 at 373; Cuthbert (1967) 86 WN (pt 1) NSW 272 at 278; Colman (Anthony Keith) NSWCCA 5 November 1997; Kukunoski NSWCCA 17 August 1989; Retton (Timothy Craig) NSWCCA 14 November 1994). Prevalence has been acknowledged as a reason for establishing a guideline in English guideline judgments. (See eg Cunningham (1993) 96 CrAppR 422 at 425; Brewster (1998) 1 CrAppR 220 at 224 per Lord Bingham).
87 “Prevalence” may refer to a situation in which a particular crime occurs with such frequency that it has a salience beyond the persons immediately affected by the crime and, accordingly, impacts on society by changing patterns of behaviour out of a sense of apprehension. None of the material presented by the Crown to this Court suggests that armed robbery is “prevalent” in this sense. Indeed the Crown submissions referred to armed robbery as “a relatively rare crime”.
88 Rather, what the Crown sought to prove was that the incidence of this particular crime had increased over recent years, so that the objective of general deterrence was entitled to greater weight than it might have hitherto received.
89 Three sources of data were presented to the Court: police statistics, court statistics and victim surveys. By reason of changes in the manner in which recorded crime statistics were kept by the NSW Bureau of Crime Statistics and Research over the years, police statistics do not permit long term trends to be analysed. The data presented to this Court was broken up into four periods: 1982-1989, 1991-1993, 1994-1995 and 1995-1997. Of the statistics for these respective periods none, save the last three year period between 1995-1997, suggests any changes of statistical significance.
90 The Crown relied on the fact that for the period 1995-1997 there was a significant increase in the number of armed robberies involving a weapon other than a firearm. The rate per 100,000 for such offences increased from 30.5 in 1996 to 53.9 in 1997 and reached 69.2 in 1998. (NSW Bureau of Crime Statistics and Research NSW Recorded Crime Statistics 1998 p41). It is unlikely that, in the case of this offence, such an increase could have been caused by an increased propensity to report the commission of offences.
91 The second source of data was the NSW Criminal Court statistics which covered the period 1991-1997, during which there does not appear to be any significant change. However these figures do show fluctuations over the years, both up and down.
92 The third source of statistics relied upon by the Crown was victim surveys. This information indicated that over the period 1990-1997 there was no evidence of any systematic upward or downward trend with respect to robbery victimisation.
93 In the light of the other material, the reported crime statistics for the years 1995-1997, which suggest a significant increase over that period, should be treated by this Court with caution when it is suggested that changes in the incidence of the crime are such as to require a response in the sentencing practice of the courts. Nevertheless, the increase over the three year period is entitled to weight when deciding whether or not a guideline judgment is appropriate and, if so, what change with respect to past sentencing practice is appropriate.
Victim Impact and Need for Guidelines
94 The second body of material to which the Crown sought to make reference concerned the impact of armed robbery on victims. This included academic literature and the report of a clinical psychologist prepared for these proceedings. This material was said to establish the objective gravity and seriousness of the offence of armed robbery. There is no doubt that impact on victims is an aspect of the seriousness of an individual offence. General patterns of impact of the character referred to in the literature to which the Crown referred, including statistical surveys, confirm the seriousness of the offence.
95 Plainly the actual impact in each particular case will vary and, appropriately, cause variations in the sentence imposed. This is not a manifestation of inconsistency. Rather, it represents the consistent application of a principle which varies in its import according to the circumstances.
96 The surveys and other literature, to which the Crown referred, establish that armed robbery is perceived as a life threatening situation for a majority of victims. In one Australian survey 86 percent reported the robbery as being the most threatening experience they had ever had. Victims experienced both physiological and psychological problems. The physical effects included chronic nervousness, insomnia, nightmares, headaches, digestive problems. Psychological problems included generalised fear, depression, aggressiveness, mood changes. Sometimes the response was such as to develop into the clinical condition of post traumatic stress disorder.
97 The surveys and academic literature confirm what trial judges and appellate judges would in any event know, namely, that armed robbery is a serious offence by reason, inter alia, of the significant impact it has on its victims. Indeed, the submissions on behalf of the Respondents to the Crown appeals did not, in contrast with criticism of the reliance by the Crown on crime statistics, challenge the materials put before the Court on the impact of the crime of armed robbery on its victims.
98 The Crown also put before the Court a report prepared for purposes of this appeal by a clinical psychologist on the effects of emotional distress in the aftermath of personal trauma, not limited to cases of armed robbery. Nothing in the report was startling or controversial and no submissions were directed to it.
99 This material confirms what this Court would in any event accept on the basis of commonsense and common knowledge. Armed robbery is not simply a crime against property. It is a crime against persons. Furthermore, the fear engendered by the perpetrator of this crime, together with the continued adverse effects on its victims, establish armed robbery to be a serious crime which requires condign punishment.
Sentencing Statistics and Need for Guidelines
100 The primary material before the court for the purpose of determining whether or not a guideline judgment is appropriate is to be found in the statistics kept by the Judicial Commission of New South Wales. Those statistics are available for all sentences imposed between December 1994 and October 1998 for offences against s97(1) of the Act, which extends to both armed robbery and robbery in company. In the period of almost four years covered by the statistics, there were a total of 835 cases. This is a significant number from which general conclusions as to sentencing practices can be drawn.
101 The most important single aspect of the statistics is that of the 835 cases, a total of 688 (eighty-two percent) resulted in a full-time custodial sentence. That is to say, 147 cases, eighteen percent of the total, did not result in a full-time custodial sentence. Of those, the largest number, 64 (eight percent) resulted in a sentence of periodic detention and 55 (seven percent) resulted in a community service order.
102 The Sentencing Information System of the Judicial Commission provides the following information for the 688 cases in which a sentence of imprisonment was imposed, divided into charts for the Full Terms and those for the Minimum or Fixed Terms.
103

At the other end of the spectrum, only seven individuals received more than half the maximum term of twenty years, i.e. more than ten years, as full terms. Furthermore, only about one-quarter of the entire sample of 688, being 163 cases, received full sentences of more than one-quarter of the maximum, i.e. more than five years.
104 The chart with respect to the minimum or fixed terms is also revealing. The statistics do not allow the removal of the cases of fixed terms in order to look at the sentencing practice with respect to the relationship between minimum and full terms. On the basis of experience, it can be said that fixed terms are comparatively few in number. The statutory formula in the Sentencing Act 1989 (NSW) is for an additional term not exceeding one-third of the minimum term, unless “special circumstances” are found to exist.
105 There are 307 cases out of the 688 total (45%) for which a minimum (or fixed) term of eighteen months or less was imposed. The application of the statutory ratio suggests that something of the order of that number of cases should have received full terms of twenty-four months or less (without making express allowance for the number of fixed terms). However, the full terms chart shows that only 73 individuals (not about 307) had sentences of twenty-four months or less imposed.
106 For the next bracket up, the figure for minimum terms is 426 individuals who received two years or less for which the statutory ratio would imply a full term of thirty-two months. The Judicial Commission’s statistics provide full terms for thirty months and thirty-six months. In the former case, there were only 106 individuals sentenced and, in the latter case, 246.
107 These statistics indicate that the “special circumstances”, which justify a ratio other than the statutory ratio, are found in the majority of cases. Furthermore, the figures strongly suggest that it is very often the case, as suggested in the particular instances before the court on this occasion, that trial judges have imposed a minimum term of significantly less than half of the full term.
108 This is an offence for which the maximum penalty is twenty years. The statistics show that of the 688 persons sentenced to prison, sixty-two percent of the total receive an actual prison term (minimum or fixed) of two years or less. Eighty percent receive actual minimum or fixed terms of three years or less.
109 These statistics indicate the following conclusions as to sentencing practices in the District Court for offences against s97(1) of the Crimes Act:
(i) Non-custodial sentences are not confined to exceptional cases. The fact that 147 individuals received non-custodial sentences out of a total of 835 (i.e. 18%) is not consistent with an “exceptional case” test.
(ii) The leniency of the sentencing pattern is also suggested in the full terms. (Over 60% of those who were in fact sentenced to prison received one fifth or less than the maximum, i.e. four years or less).
(iii) The pattern of leniency is reinforced by the extent to which the statutory ratio between minimum and fixed terms has been altered so that over 60% of those who do receive prison sentences, receive a minimum or fixed term of two years or less.
(iv) The large proportion of lenient sentences is also suggestive of inconsistency in sentencing practices between the group who receive full terms of 3-4 years or thereabouts and the group below that.
110 These statistics strongly suggest both inconsistency in sentencing practice and systematic excessive leniency in the level of sentences. They justify the promulgation of a guideline judgment.
        CCA Cases
111 One of the factors considered by the Court in Jurisic to be relevant to the determination of a need for a guideline judgment was the history of Crown appeals against sentence for the offence there under consideration, s52(A) of the Crimes Act. This Court had said, on a number of occasions, that that offence had to be treated as being more serious than it had been regarded. Notwithstanding those indications, there was a continued flow of, almost invariably successful, Crown appeals.
112 The position with respect to s97(1) in this Court is not as unequivocal as the history with respect to the offence under consideration in Jurisic. Nevertheless, some aspects of this Court’s prior consideration of Crown appeals from sentences for s97(1) convictions are relevant.
113 First, and most important, is the frequency with which the Court has stated that a non-custodial sentence for this offence could only be imposed in exceptional circumstances. As Hunt CJ at CL said in Roberts (1994) 73 A Crim R 306 at 308:
“This Court has always made it clear that armed robbery is to be regarded as an offence of the utmost gravity which, except in the most exceptional circumstances, must carry a full time custodial sentence: Murray (unreported, 11 September 1986) at p5; Kingsbeer (unreported, 29 July 1988) at p7; Valentini (1989) 46 A Crim R 23 at 26; Readman (1990) 47 A Crim R 181 at 104-105; Diamond (unreported, 18 February 1993) at p2; Hetherington (unreported, 25 February 1993) at p4; Maddocks (unreported, 25 November 1993) at pp2, 6. That view has been said by this Court to be a sentencing principle resulting from its considered decisions to which sentencing judges should not merely pay lip service, Maddocks (at p6). Wood J, delivering the principal judgment, went on to say:
        ‘I would repeat, although I wonder why it is necessary that I should do so, that it is only in the most exceptional circumstances that anything other than a non-custodial sentence should be imposed for armed robbery. Necessarily, there will be cases which constitute an exception but they will be few and far between.’
            With those observations, I express my complete agreement. I prefer the phrase “most exceptional circumstances” which he used to the phrase “wholly exceptional and unusual circumstances” subsequently employed in Crotty (unreported, NSWCCA 28 February 1994) at p5.”
114 The cases referred to by Hunt CJ at CL indicate the frequency with which this Court had reiterated this basic proposition prior to 1994.
115 The proposition, as summarised and reiterated in Roberts, has been applied subsequently in successful Crown appeals (Wright (1997) 93 ACrimR 48; Bragias NSWCCA 12 March 1997; Khoury NSWCCA 5 December 1994; Amohanga NSWCCA 25 May 1995).
116 It has also been referred to in Crown appeals in which the Court has found the first instance sentence to be manifestly inadequate but, in the exercise of its discretion, decided to dismiss the appeal. (Sharpe NSWCCA 27 September 1994; Randall NSWCCA 19 September 1994; Kerr NSWCCA 26 August 1997; Gitt NSWCCA 18 May 1998).
117 The Court has affirmed the basic proposition on occasions on which it has held that a finding of exceptional circumstances was open to the trial judge (Jones NSWCCA 15 April 1994; Georges NSWCCA 30 May 1996; Latouff NSWCCA 12 December 1996; Tocknell NSWCCA 28 May 1998).
118 To use the words of Wood J from Maddocks, which Hunt CJ at CL quoted with approval in Roberts, non-custodial sentences should be “few and far between”. One thing that is clear from the Judicial Commission’s statistics is that non-custodial sentences imposed by trial courts for this offence are common. They cannot be described as “few and far between”. This long line of authority has not been reflected in the sentencing practice of trial judges.
119 The distinction between the precedents in this Court on this offence, and the precedents considered in Jurisic, is that the Court has been prepared to uphold the finding of exceptional circumstances on a number of Crown appeals, whereas in Jurisic the disposition of appeals was almost all one way. However, prior to the present hearing, the pattern of sentencing by trial judges and this Court’s precedents on appeal, have never been reviewed in a systematic way. It is one of the advantages of a system of guideline judgments that this Court can review its own prior decisions from the perspective of ensuring consistency in the guidance it gives to trial judges. The pressures on the Court do not necessarily permit it to review its own decisions from this perspective in the normal course.
120 In the course of its prior consideration, the Court has had occasion to make reference to the sentencing statistics. In Tocknell Hulme J, with whom Hidden J and Carruthers AJ agreed, said, with reference to the conclusion that a finding of exceptional circumstances was within the sentencing judge’s discretion:
“In reaching that conclusion I place no weight on the sentencing statistics prepared by the Judicial Commission which were put before this Court. Insofar as some of those statistics suggest that something in the order of forty-seven percent of offenders guilty of the offences with which the respondent was charged and under twenty-one, had not been sentenced to full time custodial sentences, those statistics suggest to me that the judges of the District Court are not paying proper regard to the principles which this Court has laid down. If that impression be correct, and if the tendency continues, it may be that the traditional approach of this Court to appeals by the Crown may need to be reviewed.”
121 The issue arose in a more acute form in the case of Ross (NSWCCA 14 May 1997). In that case, the sentence of a two year minimum term and a two year additional term was within the trial judge’s discretion to find “the case to be an exceptional one deserving leniency”. Studdert J and Levine J both agreed with the result, but with the following observations:
“… The success of the appeal depended upon the Crown establishing that in all the circumstances the sentence was manifestly inadequate. In this regard the Crown case was not assisted by sentencing statistics produced to the court. Those statistics did not illustrate that the sentence here imposed was outside the range. Of course caution has to be used in resorting to such statistics and an appeal cannot be determined simply by reference to statistics of the nature it provided. …However, I am not persuaded that this Court should intervene.”
122 Considerations of this character are plainly material. Particular scrutiny is appropriate where statistics indicate that the sentence under appeal is at the top or bottom of the range of sentences actually imposed, in circumstances where its objective characteristics did not indicate that it is appropriately so characterised. (See e.g. Bugmy (1991) 69 CLR 525 at 538; Bloomfield (1998) 44 NSWLR 734 at 739F). However, the court in Ross did not have before it a challenge to the usual level of sentencing in the way that that issue has been presented to the Court on this occasion. The judgments in Tocknell and Ross are precursors to the present judgment.
123 The number of cases in which non-custodial sentences have been imposed indicates that the long established principle that such must be confined to “exceptional circumstances” has not been implemented by sentencing judges. The position is not quite as clear with respect to the level of the sentencing where custodial sentences are imposed. It is overwhelmingly likely that the approach to sentencing reflected in the imposition of non-custodial sentences has also been reflected in the length of custodial sentences when imposed. The sentencing statistics suggest that that is so.
124 In Waldron (NSWCCA 3 March 1994) Hunt CJ at CL said (at p3):
“This Court has on numerous occasions said that the range of sentences imposed for armed robberies, particularly serious ones, should be more deterrent than those which are in fact being imposed by the sentencing judges; see, for example Flack (12 December 1989); Smith (12 December 1984); Petrinovic (18 September 1990); Va (11 November 1993).” (3)
        See also Ellis (1993) 68 ACrimR 449 at 462; Lawson NSWCCA 12 December 1997.
125 The sentencing statistics hitherto set out, indicate that the views expressed in Waldron have not been implemented. In part, this may have occurred because of the number of cases of findings of “exceptional circumstances” which this Court has endorsed on a basis other than the exercise of its discretion. A guideline judgment is required to ensure that the pattern of leniency and inconsistency indicated in the sentencing statistics is discontinued.
Guideline Precedents
126 The determination of an appropriate sentencing guideline for offences against s97(1) must commence with a recognition that the objective and subjective factors relevant to the exercise of the sentencing discretion, may vary over a wide range. See eg Brown (1989) 17 NSWLR 472 at 473-474. As I said in Jurisic with respect to s52A:
“The nature of the offence is not such that the Court can devise a simple table in which indicative penalties are linked to a quantitative measure of the offence … What can be done, however, in case of an offence concerning a wide range of conduct which varies qualitatively rather than quantitatively, is to indicate in a general way the kind of case which would usually require a particular kind or level of sentence, whilst acknowledging that there will always be exceptional cases.” (35-36)
127 A consideration of the practice of other criminal appellate courts, with respect to guidance for sentencing for armed robbery, suggests that such an approach is appropriate for this offence. I commence with the English cases.
128 In 1972, the Court of Appeal (Criminal Division) heard together a number of severity appeals from convictions for armed robberies involving banks. The Court identified a “starting point” of fifteen years imprisonment for what the Court described as a “normal” bank robbery, where firearms were carried and no serious injury done. (Turner (1975) 61 CrAppR 67 at 91).
129 In Gould (1983) 5 CrAppR(S) 72, the Court considered a number of cases of robbery of small business premises where the offenders were armed with guns, imitation guns, or knives. The Court confirmed the guideline in Turner as applicable to a serious type of case but added:
“There are so many possible combinations of circumstance that it is difficult to give any precise indication of the so-called normal sentence for any particular type of robbery.” (75)
130 The Court went on to identify a list of mitigating and aggravating circumstances. The Court found that the sentences in the three appeals before it (between five and eight years) were not excessive. It did not specify a starting point or sentencing range.
131 The Court continued with this approach when it dealt with a number of Crown appeals in Attorney-General’s References Nos 3, 4, 8, 9, 10, 11 and 16 of 1990 (Dickson & Ors) (1991) 92 CrAppR 166. All these cases involved small shops. The Court allowed the appeals and substituted verdicts of between three and a half years and six years.
132 The Court has subsequently referred to a seven year sentence in cases of this character as if it were a starting point. The case involved a very small bank branch and an imitation gun (Attorney-General’s Reference No 14 of 1991 (Reed) (1992) 13 CrAppR (S) 446). Lord Lane CJ said:
“… it is very seldom that in these circumstances a sentence of much less than seven years would be appropriate, the purpose of the sentence being threefold: first of all to deter the offender from behaving in this way again; secondly to deter others from arming themselves with weapons and holding up banks; and thirdly, a matter which is sometimes overlooked, to punish the offender for having carried out this wicked crime and having put several people in fear of death.” (449)
133 The seven year sentence in Reed was referred to as a starting point by Lord Taylor CJ in Attorney-General’s Reference No 7 of 1992 (Khan) (1993) 14 CrAppR (S) 122 at 125, 126, which also involved a small shop and an imitation pistol. His Lordship added:
“It has to be realised that that type of shop is very often staffed by only one person who may be unable to defend himself or herself. It is unlikely that there will be any sophisticated security there, and it is a primary target for someone who wants to enrich himself quickly and successfully. It is therefore very important that the courts should indicate by the sentences passed that that type of offence will be punished severely.” (126)
134 I should note that the Court of Appeal has subsequently issued a separate guideline judgment with respect to the use of firearms in a variety of offences (Avis (1998) 1 CrAppR 420).
135 This reluctance in Gould to specify a range or starting point was based on the wide variation in the factual circumstances of individual offences. The reluctance to offer any guidance, where complete guidance is not feasible, has been criticised. (Ashworth “Techniques of Guidance on Sentencing” (1984) CrimLR 519 at 529-530; Ashworth Sentencing and Criminal Justice (2nd ed, 1995) at 30. It may be that that subsequent decisions are a response to this criticism.
136 Guidance may be given by identifying certain commonly recurring categories of a specific offence. The New Zealand Court of Appeal has done this in Moananui (1983) NZLR 537. Guideline judgments in New Zealand are “bottom up” not “top down” guidelines, i.e. they purport to describe, rather than prescribe, sentencing practice. (See Hall “Reducing Disparity by Judicial Regulation: Sentencing Factors and Guideline Judgments” (1991) 14 NZ ULR 208 esp at 223-224).
137 In Moananui the Court identified three categories:
        (i) Planned armed robbery at premises such as banks which endanger the safety of considerable numbers of people: which usually attract sentences of 6-8 years.
        (ii) armed robbery involving intrusion into dwelling houses, usually at night: examples given range from 4-6 years.
        (iii) armed robberies of smaller premises involving less extensive risk of injury and smaller sums of money: the range of sentences is lower. (Examples range from 2-5 years).
138 A more overtly structured approach has been adopted in Hong Kong. In Mo Kwong Sang (1981) HKLR 610, the Court identified what it described as an “ordinary case” to be a starting point: armed robbery by an offender carrying a knife or other dangerous weapon, which was displayed to the victim. The Court said that an “appropriate” sentence for this case was five years. It indicated appropriate levels for two other common factual variations: invasion of private premises (six years) and physical violence (seven years).
139 The Court also identified a list of aggravating and mitigating factors, noted that use of firearms would require more severe sentences, and affirmed the discretion of the sentencing judge to take into account subjective circumstances. The nature of the guidance provided by the appellate court was to:
“… indicate the level of sentence for armed robbery with which we would not interfere. (611)
140 I have noted above that Canadian criminal appellate courts have adopted a “starting point” approach to providing guidance to trial judges. This involves the identification of a typical case and the application of aggravating and mitigating factors. Courts in Alberta, Nova Scotia and New Brunswick have applied this approach to armed robbery. (See Young The Role of an Appellate Court in Developing Sentencing Guidelines Canadian Sentencing Commission (1988) p21ff).
141 In Nova Scotia the appellate court has lain down a three year minimum sentence (Brennan and Jensen (1975) 11 NSR (2d) 84 at 88; Hingley (1977) NSR (2d) 541 at 544). The New Brunswick Court of Appeal has referred to three years as a normal sentence (Chaisson (1975) 24 CCC (2d) 159). In both cases the sentencing level was identified as appropriate to a young first offender, but there was little in the way of precision about other aspects of the offence or of the offender.
142 In Johnas (1982) 2 CCC (3d) 490 the Alberta Court of Appeal focused on cases of robberies of “small commercial establishments, open at night for service to and convenience of the public, where often a single person is in charge” (494). The objective circumstances emphasised the degree of vulnerability. The category identified involved a threat of violence, but no actual violence. The Court identified the category as follows:
“… what should be regarded as a fit sentence in Alberta for unsophisticated armed robbery of unprotected commercial outlets in the absence of actual physical harm to the victim and with modest or no success.” (495)
143 The Court determined three years as a “starting point” and added:
“… judicial reasoning as to a fit sentence for any offence must start with a norm for the type of offence involved. That norm is arrived at by comparisons to other cases, by experience, by the seriousness of the offence and by its prevalence. Having determined that norm, the court will look at the factors of mitigation and aggravation. Specific cases are not to be treated as precedents. Each is a unique mixture of aggravating and mitigating factors.” (499)
144 In subsequent cases the Alberta Court of Appeal has specified five years as a starting point for bank robberies (Kurich (1982) 9 WCB 138) and four years for night deposit robberies (Hall (1983) 10 WCB 138).
145 The Court of Criminal Appeal of Western Australia has, with respect to armed robbery, provided sentencing guidance in, what I have called above, a “bottom up” fashion: i.e. it is derived from the range of sentences actually imposed by trial judges. This involves identifying a tariff rather than fixing a starting point or sentencing range. It has been called “tariff sentencing”. (Thomas Principles of Sentencing (2nd ed, 1979) p29).
146 In 1989, what the Court described as a “conventional armed robbery of a bank or similar premises”, attracted a sentence of “five to seven years or upwards” (Norman WACCA 1 February 1989). By 1997 these levels had been increased. The Court said:
“… sentences have tended to firm up as a result of the increasing prevalence of the offence by giving greater weight to the requirements of deterrence and less weight to the antecedents and other matters personal to the offender … The offence of armed robbery has become significantly more prevalent since 1989 and sentences have firmed up in that period. At present, the range of sentences commonly imposed for a single offence, depending on the circumstances, would be from six to nine years.” (Miles (1997) 17 WAR 518 at 521 and 522).
147 This level applies before a discount for a plea of guilty. The Court went on to note that where a blood filled syringe was used as a weapon, an additional one year’s imprisonment should normally be imposed, because of the additional impact on the victim (523-524).
148 The Western Australia Court of Criminal Appeal has affirmed the range of 6-9 years in subsequent decisions (Moody 12 February 1998; Moulds 4 March 1998; Jeffree 1 May 1998). In Jeffree, the Court further clarified the application of the range which was described in Miles as “commonly imposed”:
“Offences of this type commonly do not progress beyond a threat or violence of a minor kind. Those which do could be characterised as amongst the worst type of case and would be outside the usual range.”
149 The range of sentences for this offence in Western Australia - six to nine years - must be understood in the light of a discount for a guilty plea (said to be 20-35%, Miles at 521) and the statutory framework for sentencing, which differs from that in New South Wales. By s93 of the Sentencing Act 1995 (WA), the range of six to nine years could permit release on parole after, in broad terms, two to four years. This, again in broad terms, would equate to a minimum term under the Sentencing Act 1989 (NSW).
150 As I have indicated above, the Full Court of the Supreme Court of South Australia has promulgated “sentencing standards” in certain cases. In Cadd (1997) 94 ACrimR 466 at 480 Doyle CJ identified the authorities which had set such standards for the offence of armed robbery. He referred to Spiero (1979) 22 SASR 543 as setting a standard and to Dube (1987) 46 SASR 118 as a case in which the Court contemplated, but refrained from, increasing the level.
151 In Dube the Court considered a Crown submission that this standard should be increased because of the increased prevalence of the offence. The submissions made are strikingly similar to those before this Court in the seven cases now before it. King CJ noted:
“The prosecution adduced evidence before the learned sentencing judge as to the adverse affect of armed robbery upon the victims and innocent people who become embroiled in it. There was also evidence as to the financial cost of armed robberies to financial institutions and to the community generally … The evidence led by the prosecution in this case merely reinforces the view long held on this Court that armed robbery presents a serious threat to the wellbeing of the community and that punishments imposed must reflect that fact …
        The prosecution produced evidence before the learned sentencing judge of the increasing prevalence of this crime during the last seven years. It is unnecessary to recount the details of this evidence; it is sufficient to say that it establishes that the increase has been marked and is apparently continuing. The much discussed question of the effectiveness of imprisonment as a deterrent to crime, and in particular of the effectiveness of increased levels of punishment, was adverted to during argument. I think that it must be conceded that there is no proven correlation between the level of punishment and the incidence of crime and that there is no clear evidence that increased levels of punishment have any effect upon the prevalence of crime. Nevertheless the criminal justice system has always proceeded upon the assumption that punishment deters and that the proper response to increased prevalence of crime of a particular type is to increase the level of punishment for that crime. I think that courts have to make the assumption that the punishments which they impose operate as a deterrent. That being so, I think that it follows that the proper response, and the response which is expected by the community at large, to the increased prevalence of serious crime is increased punishment for that crime.” (119-120)
152 The reason the Court refrained from formally increasing the standard was because of the then recently amended legislative provision on the treatment of remissions, which would have the effect of increasing sentences.
153 Spiero involved an armed robbery of a chemist shop by three offenders. Firearms were used. The sentence of eight years fixed in that case was treated subsequently as a standard. The appropriate range has been said to be eight to twelve years, in all but “exceptional cases”. (See Fermaner (1994) 72 ACrimR 138 at 139; 61 SASR 447 at 448 based on Prendergast (1988) 147 LSJS 486 at 486-488).
154 Fermaner was heard together with three other Crown appeals from sentences for the offence of armed robbery: Bini, Meo and Ramsden, each SACCA 21 March 1994 (unreported). The four cases are analysed in Hinton “Principled Sentencing in South Australia - Armed Robbery and the Quest for Consistency” (1995) Flinders J L.Reform 18.
155 Three of the four cases have similar objective circumstances: robbery of a service station; offenders aged from 24 to 36 years; armed with a knife, a screwdriver and a small axe; a few hundred dollars taken; an early plea of guilty. There were differences in terms of prior record and effect on victim. The Court of Criminal Appeal set aside the sentences below and imposed sentences of five years four months (on top of five months served before sentence); with a non-parole period of four years; six years with a non-parole period of four years and 5 years 8 months, with a non-parole period of three years.
156 In these cases (Fermaner, Bini and Ramdsen) eight years was treated as a starting point. In Fermaner the Court indicated that this would have been the appropriate head sentence but for the early plea of guilty and the period of five months in custody before sentence (140). This was reiterated in Bini and Ramsden. The plea of guilty led to a discount of two years (Hinton supra p27).
157 Since the reaffirmation of the eight year starting point in these four 1994 cases, a system of truth in sentencing was introduced by the Statutes Amendment (Truth in Sentencing) Act 1994 (SA). As a result, the accrual of remissions, which the Court had referred to in Dube, would no longer be taken into account in sentencing. This would require a reduction of the appropriate range hitherto identified for armed robbery by the extent to which sentences had been inflated to take account of remissions. A reduction of one third would accord with the transitional provisions of the Statutes Amendment (Truth in Sentencing) Act (See Hinton supra pp25-26).
158 This suggested a reduction in the eight to twelve years for the head sentence, to about six to eight years. This has since been affirmed by the South Australian Court of Criminal Appeal. (Drumgoon SACCA 20 November 1995; Branscherd SACCA 22 May 1996). The starting point of six years in South Australia was before a plea of guilty and other discounts (eg co-operation with the police). (See Drumgoon supra).
159 The Queensland Court of Appeal reviewed a number of prior decisions in Hammond (1996) 92 ACrimR 450 at 456ff. The case involved pleas of guilty to three robberies of service stations by an offender armed with a steering wheel lock. After referring to the need for consistency with the prior decisions it reviewed, the Court imposed a sentence of five years with a recommendation for release on parole after two years.
160 In a subsequent case involving a single armed robbery of a service station when armed with a knife, by an offender who pleaded guilty, the review in Hammond was referred to, the severity appeal allowed and a sentence of four years substituted. (Morton (1997) 95 ACrimR 381).
A Guideline for New South Wales
161 In the cases now before this Court, the Crown has propounded a starting point for New South Wales of six years full term, in circumstances where there is a plea of guilty.
162 It appears from the cases that come to this Court, including the present proceedings, that there is a category of case which is sufficiently common for purposes of determining a guideline:
        (i) Young offender with no or little criminal history
        (ii) Weapon like a knife, capable of killing or inflicting serious injury
        (iii) Limited degree of planning
        (iv) Limited, if any, actual violence but a real threat thereof
        (v) Victim in a vulnerable position such as a shopkeeper or taxi driver
        (vi) Small amount taken
        (vii) Plea of guilty, the significance of which is limited by a strong Crown case.
163 Whilst it is possible to determine a starting point in a case of this kind, i.e. a sentence of X years imprisonment, I do not believe that the Court should do so. Rather, I propose the Court should identify a narrow sentencing range within which this Court would expect sentences in such cases to fall.
164 There are two principal reasons why a sentencing range is appropriate for this offence:
        (i) The seven characteristics identified above do not represent the full range of factors relevant to the sentencing exercise.
        (ii) Many of the seven identified characteristics contain within themselves an inherent variability, eg different kinds of knives or weapons in (ii); extent of “limited actual violence” in (iv); degree of vulnerability in (v); amount in (vi).
165 In my opinion sentences for an offence of the character identified above should generally fall between four and five years for the full term. I have arrived at this figure after drawing on the collective knowledge of the other four members of the Court with respect to sentence ranges. I have also reviewed the sentences which this Court has imposed on occasions when it has intervened, including in Crown appeals where the principle of double jeopardy applies. The proposed range is broadly consistent with this body of prior decisions in this Court.
166 Cases consistent with the range are Sandaford 10 September 1995; Sedgwick 7 July 1992; Martin 19 February 1993; Sarkas 8 September 1992; Walters and Watkins 2 June 1994; Achurch & Brady 8 July 1994; Willett 6 November 1995; Maretta 30 October 1996; Bragias 12 March 1997; Basaga & Weleilakeba 9 August 1998.
167 Cases above the range to varying degrees are Salameh 12 March 1991; Sneddon 28 March 1991; Marsden 15 April 1993; Antoce 22 April 1993; Walker 21 February 1994; Pettit 9 March 1994; Sneddon 28 March 1994; Nguyen 14 April 1994, Burger 19 July 1994; Rafter 23 September 1994; Harborne 12 October 1994; Murray 7 November 1995; ; Putescu 20 December 1996; White 7 April 1997; Lowe 24 April 1997.
168 Cases below the range being sentences between 2 and 3½ years are Bateman 15 July 1993; Roberts 24 May 1994; Amohanga 25 May 1995; Davies 29 November 1995; Wright 28 February 1997; Fisher 8 September 1998; Wightman 2 November 1998.
169 Aggravating and mitigating factors will justify a sentence below or above the range, as this Court’s prior decisions indicate. The narrow range is a starting point.
170 In addition to factors which may arise in any case eg youth, offender’s criminal record, cooperation with authorities, guilty plea in the absence of a strong case, rehabilitation efforts, offence committed whilst on bail etc, a number of circumstances are particular to the offence of armed robbery. These include:
        (i) Nature of the weapon
        (ii) Vulnerability of the victim
        (iii) Position on a scale of impulsiveness/planning
        (iv) Intensity of threat, or actual use, of force
        (v) Number of offenders
        (vi) Amount taken
        (vii) Effect on victim(s)
171 The Respondents to the Crown Appeals also submitted that this Court should now declare that the fact that the offender was addicted to drugs, and committed the crime to obtain money to feed his or her habit, is a mitigating circumstance. This would require the Court to overturn a long line of prior decisions.
172 In 1990, the Australian Institute of Criminology published a detailed analysis of the practice of this Court with respect to sentencing for robbery over the course of a ten year period. That report made it clear that the Court had consistently regarded drug addiction as a mere explanation, not an excuse. (Potas Sentencing Robbers in New South Wales (1990) at pp142, 163).
173 As a joint judgment of the Court put it in Valentini (1989) 46 ACrimR 23 at 25:
“This Court has said on countless occasions that addiction to heroin is not to be considered as a factor for the reduction of what would otherwise be an appropriate sentence for the nature of the offences which have been committed. It serves, however, to provide an explanation for the commission of the offences.”
        This has been reiterated in a number of later cases (eg Stanford NSWCCA 23 November 1993; Atkins NSWCCA 27 May 1998).
174 Drug addiction is one of the circumstances of a particular offence that is relevant to the sentencing exercise. It may, for example, be pertinent to the issue of impulsiveness/planning or to the weight to be given to rehabilitation in a particular case.
175 In Engert (1995) 84 ACrimR 67, Gleeson CJ emphasised the complexity of the sentencing task and the multiplicity of the, sometimes conflicting, objectives it must serve. His Honour concluded:
“It is therefore erroneous in principle to approach the law of sentencing as though automatic consequences follow from the presence or absence of particular factual circumstances. In every case, what is called for is the making of a discretionary decision in the light of the circumstances of the individual case, and in the light of the purposes to be served by the sentencing exercise.” (68)
176 His Honour went on to say that the existence of a causal relationship between a mental disorder and the commission of an offence does not automatically produce the result that the offender will receive a lesser sentence (at 71). Nor, in my opinion, should the existence of a causal relationship between drug addiction and the commission of an offence.
177 The circumstances in which motive may be a mitigating factor should, in my opinion, be confined to cases in which motive impinges upon the moral culpability of the offender. This can include mental, emotional or medical problems or impulsive conduct. In any event the degree to which motive may be pertinent must depend on the whole of the circumstances. (White NSWCA 23 June 1998 at 24).
178 As noted above, it has long been the position taken in this State that addiction is not, of itself, a mitigating circumstance. This is also the position in England, (Brewster (1998) 1 CrAppR 220 at 226) Canada (Johnas (1982) 2 CCC (3d) 490 at 497). It was submitted for the Respondents to the Crown appeals that authorities in other Australian states indicate a different approach.
179 Reliance was placed on Nolan (1998) VICCA 135 (2 December 1998). This was a majority decision to allow a severity appal. One member of the majority, Buchanan JA, referred to drug addiction in terms of “mitigating circumstances” (par 15). The other member of the majority, Butt JA, did not adopt that terminology, but rather that of “explanation” (par 19).
180 Buchanan JA referred without disapproval to Halewyn (1984) 12 ACrimR 202 at 203 where Young CJ said:
“… it has been said a number of times in this Court that the fact that an armed robbery is committed to support a drug habit is of little consequence in mitigating the sentence to be imposed.”
181 Buchanan JA also referred to Bouchard (1996) 84 ACrimR 499 where at 501-502, Callaway JA quoted Hallwyn but identified a case of “a man crazed by a drug craving who committed an offence on the spur of the moment and later repented”.
182 Nolan was such a case. It should be understood as an authority on the issue of planning/impulsiveness, and that the impulsiveness was the “mitigating circumstance”, rather than the drug addiction itself.
183 I do not understand there to be any difference between the position in New South Wales and that in Victoria in this regard.
184 In Douglas (1995) 56 FCR 465 the Full Court of the Federal Court heard an appeal from the Supreme Court for the Australian Capital Territory. In a joint judgment, Von Doussa, Higgins and R D Nicholson JJ said (at 470):
“It is, of course, not a mitigating factor that a person commits a crime to feed a drug addiction. However, that is not a principle of universal application. As Jenkinson J noted, at 105, in Talbot (1992) 34 FCR 100, the moral and legal fault for the acquisition of the addiction has to be considered:
        ‘… evaluation of moral culpability remains in my opinion as fundamental to one system of punishment for crime as evaluation of the degree of harm caused or threatened by the criminal act or actor.’
            The age of the offender when he or she becomes addicted and the degree of judgment open to them at that age is thus relevant in evaluating the extent to which they should be punished for consequential criminal conduct.”
185 Douglas involved an addiction that commenced at age eight and was total by age 13. This is an extreme case of little assistance in establishing a general principle that addiction lessens moral culpability. On the contrary, it affirms that self induced addiction at an age of rational choice establishes moral culpability for the predictable consequences of that choice. That was the very distinction made by the Full Court in Talbot (1992) 34 FCR 100 at 105-106.
186 In that case, the Court took into account an addiction that was caused by medical treatment “to which his consent was at best merely formal” (105). (See also the Victorian decision in Redenbach (1990) 52 ACrimR 95 at 99). The Court referred with approval to, whilst distinguishing, the reasons of King CJ in Spiero (1979) 22 SASR 543:
“One feels sympathy for the person who has become entangled in drug addiction, but the courts cannot treat addiction as an excuse, or even a mitigating factor, in relation to serious crime. Those who are addicted to drugs must understand that if they allow their addiction to lead them into serious crime, they must expect to receive the same serious punishment as would be received by others.” (549)
187 In Terizakis (1986) 41 SASR 252 at 256, the Full Court reiterated this reasoning. O’Loughlin J added:
“Despite the cravings caused by the addiction, the majority of drug addicted persons still retain, in varying measures, the ability to distinguish between right and wrong and the ability to comprehend that the consequence of armed robbery will be a lengthy term of imprisonment. To give way, out of sympathy and concern, by imposing lesser terms of imprisonment that would otherwise be imposed, might appear to be an act of humanity, but it will do nothing to stamp out the drug trade and worse, it might even give some indirect encouragement to would-be offenders.” (256)
188 Another authority on which the Respondents to the Crown appeals rely is the decision of the Queensland Court of Appeal in Hammond (1996) 92 ACrimR 450.
189 In Rosenberger (1996) 76 ACrimR, that Court had indicated that intoxication, whether by alcohol or drugs, would not mitigate penalty save in the case where the original addiction did not involve a free choice, (as in Talbot or Redenbach supra, the latter being quoted in Rosenberger).
190 In Hammond the Court distinguished Rosenberger and said:
“The true relevance of drug addiction as a factor contributing to the commission of crime and its effect in the sentencing process has never been adequately explained. In our view it is a factor that may help an offender to the extent of showing that his or her descent into the crime in question was a secondary consequence of desperation produced by a human weakness rather than a primary choice. It may be inferred in many cases (and the present case is very much in point) that the offender would almost certainly never have become a thief but for loss of control of an ordered life through drug addiction. The proposition that the results of drug addiction are self-inflicted is half true and therefore dangerous. The offender must of course be held responsible for his or her own actions. The drug addiction is not an excuse; but it is a factor that may tell the court that the real weakness of character is that of a drug addict rather than that of a robber. That may be by no means inconsequential. It is however a two-edged factor; it may also tell the court that rehabilitation is going to be difficult.
        Just how these factors are to be applied to particular cases had not been spelled out, and it may be undesirable to attempt to do so. In our view the fact that drug addiction has led to criminal activity can properly lead to a wider understanding of the offender’s character and motivation for the crime. Depending on the circumstances of the particular case this may sometimes show the offender to be less deserving of condemnation than the primary facts, without more, might suggest.” (455-456)
191 In its conclusion the Court returned to this theme:
“Addiction to drugs at the time of the offence is not an excuse but may tell the court that the primary weakness of character is that of an addict rather than that of a robber. Such a factor tends to be mentioned along with an unfortunate family life as an explanation of the offender’s descent into bad habits. When the original problem was not something for which the offender was responsible, it is natural to feel some sympathy for the offender, though this must not obscure the offender’s direct responsibility for the ultimate criminal act he or she chose to do.” (467)
192 This latter passage appears in the context of the Court listing aggravating and mitigating factors. It is noticeable that addiction is not identified as a mitigating factor. In neither passage is addiction, of itself, treated as a basis for imposing a lower sentence than would otherwise be imposed. In both passages it is emphasised that addiction is “not an excuse”. In both passages the offender’s responsibility for his or her conduct is mentioned.
193 In my opinion, Hammond affirms that drug addiction is a relevant circumstance but is not, of itself, a mitigating factor. If, contrary to this conclusion, the remarks in Hamilton were intended to suggest otherwise - and some of the language chosen may be so interpreted - then they should be regarded as anomalous and should not be followed.
194 The authorities are against the proposition that drug addiction should, of itself, be accepted as a mitigating factor. There is authority that where the original addiction was not a willed act, that may be taken into account by way of mitigation. The authority does not go beyond that.
195 It is in the context of this line of authority that the submissions made for Respondents to the Crown appeals to the effect drug addiction, at least where it can be shown to be causally related to the commission of an offence, should now be accepted to be a mitigating circumstance.
196 It was submitted that the degree of moral culpability of a particular offender is diminished by addiction. Evidence was put before the Court that there is, at least in some cases, a genetic predisposition to addiction and that addiction generally is not simply a state of mind but has a neurobiological or physiological base. It was put that an addict’s decision to perform a criminal act was not “a completely free choice”.
197 In my opinion drug addicts who commit crime should not be added to the list of victims. Their degree of moral culpability will vary, just as it varies for individuals who are not affected by addiction. There are a number of aspects of the relationship between drug addiction and crime which indicate that moral choices are made.
198 First is the original decision to experiment with drugs which, in the usual case, is a completely free choice. The addictive quality of drugs, together with the anti-social behaviour which so commonly results from addiction, is so widely known that persons who choose a course of addiction must be treated as choosing its consequences.
199 Secondly, the submissions in this Court were in error in identifying the relevant conduct as the craving associated with withdrawal. The material presented to the Court did not suggest that the choice faced by addicts was between this negative feeling and the need for money to allay it. Rather the choice may often be the desire for the positive feeling said to be associated with a drug induced euphoria. The desire to bring about that state of “well-being” is, relevantly, a moral choice.
200 Thirdly, nothing in either the process of addiction or its neuro-biological and physiological basis, leads ineluctably to the commission of crime, let alone the commission of crimes of violence against persons, such as armed robbery. Not all persons who suffer from addiction behave in this way. Those that do so, make a choice.
201 Finally, individuals do emerge from addiction. They do so with difficulty and generally need significant amounts of help. The decision to persist with an addiction, rather than to seek assistance, is also a choice.
202 There is no warrant, in my opinion, to assess a crime induced by a need for funds to feed a drug addiction, as being lower in the scale of moral culpability than other perceived requirements for money.
203 Counsel making these submissions was driven to accept the proposition that an addiction to gambling, with its attendant need for money, could similarly be the basis for a claim for mitigation. He went further and indicated that an elevated sex drive which also had a physiological basis, could be a mitigating factor for the commission of a rape or for the conduct of a paedophile. In my opinion all these submissions should be rejected.
204 It was also submitted that the principle of general deterrence should be given less weight in the context of offenders who commit their crimes for the purposes of assuaging a drug addiction. In this submission, reliance was placed on the often expressed doubts about the direct effect on potential offenders of increases in penalties imposed. Particular reliance was placed on the driven nature of a drug addicts, many of whom engage in high risk activity, like exchange of needles notwithstanding the risk of HIV infection. This kind of submission has been made many times before and I have quoted a few instances above. It has always been rejected. This Court should do so again.
205 General deterrence always operates at the margin. Some people will continue to engage in criminal conduct notwithstanding the level of, or increases in the level of, the penalties they suffer. However, some people will be deterred. It is not to the point that some addicts engage in high risk activities. It would be necessary to establish that all addicts do so. Neither the submissions, nor the materials in support, suggest anything of this character.
206 I attach particular significance to the impact that acknowledgment of drug addiction as a mitigating factor would have on drug use in the community. The sentencing practices of the courts are part of the anti-drug message, which the community as a whole has indicated that it wishes to give to actual and potential users of illegal drugs. Accepting drug addiction as a mitigating factor for the commission of crimes of violence would significantly attenuate that message. The concept that committing crimes in order to obtain monies to buy an illegal substance is in some way less deserving of punishment than the commission of the same crime for the obtaining of monies for some other, but legal, purpose is perverse.
207 It may very well be the fact that increased possibility of detection has greater effect by way of deterrence than increased punishment. There is no warrant, however, for the Courts abandoning reliance on the latter. In any event the two propositions are related. It is only because detection, when it occurs, leads to a level of punishment, that increases in detection have their deterrent effect.
208 It may very well be that the criminal justice system has a modest role to play in the control of drug addiction. But however modest that role may be, it must be performed in accordance with the basic structure of the criminal sentencing process. At the level of a structure deeply embedded in our society, not merely at the level of an individual’s calculus of risks and benefits for specific conduct, the criminal justice system is now, and has always been, based on the proposition that punishment deters and, within limits of tolerance, increased punishment has a corresponding effect by way of deterrence. This Court should not change such a longstanding assumption. Legislation would be required to alter the common law in this way.
209 I reiterate that the process of imposing penalties for the commission of crimes, has its primary deterrent effect through its operation as a structural phenomenon of the criminal justice system. That is not capable of being assessed from the perspective of what particular penalties, or increases in penalty, may have in the case of individuals.
210 In any event, the reasons for the guideline propounded in this judgment do not relate merely to an increase in the size of penalty. The guideline is particularly directed to overcoming the very significant proportion of cases in which non custodial sentences have been imposed. Henceforth, such sentences should be restricted to the exceptional cases to which the authorities have always referred.
211 The deterrent effect of a sharp reduction in the proportion of non-custodial sentences which, as a result of this judgment, become custodial sentences, may well be much more significant than the deterrent effect from an increase in the level of custodial sentences. That is not to say that the latter does not have some deterrent effect but the sentencing practices, which I have analysed above, have been such as to significantly attenuate the deterrent effect of sentencing for the offence of armed robbery in the past. It is my opinion that that should change.
212 I have now read the additional observations of Wood CJ at CL. I agree with his Honour’s judgment.
**********


IN THE COURT OF
CRIMINAL APPEAL

No. 60559/98
60558/98
60561/98
60746/98
60596/98
60595/98
60511/98
SPIGELMAN CJ
WOOD CJ at CL
NEWMAN J
HULME J
SIMPSON J
    Wednesday 12 May 1999
      REGINA v Paul Anthony HENRY
      REGINA v Stephen Anthony BARBER
      REGINA v Hoai Vinh TRAN
      REGINA v Troy David SILVER
      REGINA v Theo TSOUKATOS
      REGINA v Bill KYROGLOU
      REGINA v John David JENKINS
      GUIDELINE JUDGMENT


      213 WOOD CJ at CL: I have read in draft the reasons of Spigelman CJ. I agree with the reasons for delivering a guideline judgment in relation to the offences before the Court, and with the guidelines proposed.
      214 I wish, however, to examine in a little more detail the submission that this Court should now declare that the fact that an offender was addicted to drugs, and committed an armed robbery in order to obtain money to support a drug habit, is a mitigating circumstance. As Spigelman CJ has observed, this would require the Court to overturn a long line of prior decisions, and to part from the position adopted in other jurisdictions, both within this country and elsewhere.
      Drug Addiction
      215 It cannot be gainsaid that very many offences of armed robbery are committed because of an addiction to drugs. All of the respondents to the Crown appeals presently before the Court were motivated by a drug habit. The New South Wales Bureau of Crime Statistics and Research reported in 1987 that:
      “drug (usually heroin) dependency appears to be an important ingredient in robbery offending, and in particular in the pattern of offending of those with multiple robbery convictions.”
      (New South Wales Bureau of Crime Statistics and Research, Robbery: Final Report (1987) at 88. When the Bureau interviewed convicted armed robbers, it found that drug use was a motivating factor in 70% of robberies (ibid at 93.)
      216 The acceptance of the link between drug dependency and many forms of criminal activity has recently led the State of New South Wales to create a Drug Court, with a special regime for dealing with certain categories of offenders, although not those charged with offences “involving violent conduct”: Drug Court Act 1998 (NSW) S 5(2)(b).
      217 The question whether drug addiction should operate as a mitigating factor when sentencing an offender for an offence under S 97(1) Crimes Act, or for similar offences, has been considered by this Court upon numerous occasions. The attitude taken to date is accurately reflected in Valentini (1989) 46 A Crim R 23; and in Ellis (1993) 68 A Crim R 449.
      218 In Valentini, the Court (comprised by Maxwell, Carruthers and Loveday JJ) said, (at 25):
      “This Court has said on countless occasions that addiction to heroin is not to be considered as a factor for the reduction of what would otherwise be an appropriate sentence for the nature of the offences which have been committed. It serves, however, to provide an explanation for the commission of the offences.”
      219 In Ellis, Hunt CJ at CL said (at 462):
      “The maximum sentence enacted by the legislature indicates that it expects the courts to impose severe sentences for these offences. I believe that the community has the same expectation. As was said by this Court in a bank robbery case (Petrinovic (unreported, 18 September 1990) at 9) the citizens of this community are sick and tired of armed robberies conducted by criminals in order to feed their drug habits, and this Court will support judges who recognise these as serious crimes and who impose heavy sentences for them.”
      220 Although Kirby P (at 459) expressed his concern in Ellis at the possibility of the Court being diverted from its legal duty, by “appeals to populist notions of what the Court conceives the community to be ‘sick and tired of’”, and dissented in the decision of the majority in dismissing the appeal, that was attributable to the comparison that he thought appropriate with the sentences that had been imposed upon separate offenders in analogously similar cases. His Honour did advert to the fact that:
      “people addicted to illegal drugs frequently have had resort to armed robberies in an attempt to secure the funds to feed the habit made expensive by its illegality”
          but his judgment provides no encouragement for the view that such an addiction should be regarded as a factor in mitigation.
      221 In Begnell (Court of Criminal Appeal New South Wales 26 November 1992) his Honour observed (at 4):
      “The courts cannot, in the face of the provisions of the statutes which criminalise the conduct of the use of prohibited drugs, allow that use to become an excuse for the kind of conduct which results in the charges and convictions that bring the applicant before us today. On the other hand, it is necessary, as it seems to me, to take into account such a consideration in looking at the criminality of the action in its totality and in considering its seriousness.”
      222 I do not understand his Honour, by the qualification contained in the second sentence of this passage, to be suggesting that the fact that an offence was committed to cater for a drug habit, should be regarded as a factor of mitigation. The contrary was, in fact, suggested by his Honour’s comments in Bradley (Court of Criminal Appeal New South Wales 26 October 1993) in a case of assault and rob involving a bag snatch from an elderly lady in a suburban street by an offender with a cocaine dependence, (at 5):
      “… her past addiction to cocaine was not an excuse for the action of the kind which occurred on this occasion. There are organisations and individuals in the community who will help people with this form of addiction. It would be completely intolerable if people with an addiction of this kind could resort to acts of violence against elderly members of the community or indeed anyone else. The courts must make it plain that such conduct will not be tolerated.”
      223 Similarly, in Vidler (Court of Criminal Appeal New South Wales 10 April 1986) Street CJ, with whom Lee and Lusher JJ agreed, said that:
      “Drug addiction with the consequently generated cash need … may explain criminal activities but it cannot excuse them so as to result in diminution in sentences proper to be passed for the offences under consideration”.
      224 In Crotty (Court of Criminal Appeal New South Wales 28 February 1994) where the offence involved was one of armed robbery of a taxi-driver, Sully J said (at p6):
      “Drug induced crime, particularly drug induced armed robbery, will not be regarded by the Courts as being necessarily mitigated by the particular drug and alcohol problem. It needs to be said yet again that ordinary decent members of the community who look, and rightly look, to the Courts for protection of their persons and property against assault and robbery are fed up with the notion that somehow or other a person who assaults them, who steals their property, who breaks into and ransacks their homes, is somehow not to be blamed for what he or she has done because of the presence of some drug or alcohol or like sociological problem invoked to explain what has been done in a particular case. I do not suggest, of course, that the presence of a drug or alcohol problem is not a relevant consideration.
          All I am seeking to say is that, as with the bare fact of youth, so also with the bare fact of drug and alcohol inducement, it must not be allowed to become a cloak of convenience behind which people who are in every relevant sense adults can shelter from the responsibility for their serious breaches of criminal law.”
      225 More recently it was said, by McInerney J, with whom Bruce J agreed, in Jebara (Court of Criminal Appeal New South Wales 2 August 1995):
      “It has been pointed out that whatever sympathy the courts may feel for a person who is subject to a compulsive and expensive habit, it cannot be made an excuse for crime. The community has to be protected against criminal action, whatever the motivation may be. In R v Martin (Court of Criminal Appeal New South Wales 19 March 1992 unreported) Hunt CJ at CL referring to such a habit providing a reason for the commission of offences, stated it provided no excuse in mitigation. At best, such a circumstance avoids a finding of aggravation by reason of the offence being committed for pure greed."
      226 This approach, which has been referred to as a “hard line” approach (Potas, Sentencing Robbers in New South Wales 1990 at 242, 163 can also be seen in decisions such as: Atkins (Court of Criminal Appeal New South Wales 27 May 1999); Hines (Court of Criminal Appeal New South Wales 25 May1998); Salameh (Court of Criminal Appeal New South Wales 9 June 1994); Crotty (Court of Criminal Appeal New South Wales 28 February 1994 unreported); Stanford (Court of Criminal Appeal New South Wales 23 November 1993); Jansz (Court of Criminal Appeal New South Wales 30 September 1993); Shinner (Court of Criminal Appeal New South Wales 10 July, 1992); and Cottier (Court of Criminal Appeal New South Wales 18 December 1990).
      227 A departure from this line of authority can be seen in the observation of Hulme J, in Murray (Court of Criminal Appeal New South Wales 7 November 1995) where his Honour said that:
      “the fact that (the applicant’s second criminal) career was inspired by a need for funds to support his heroin habit, is, to my mind, a mitigating factor”.
      228 He was, however, in the minority on this point, Grove J, with whom Hunt CJ at CL agreed, saying:
      “I would not regard an explanation founded in a need to support a heroin habit as a mitigating factor in regard to these offences. No doubt the habit offers explanation but in my view, it does no more than that.”
      229 A similar stand has been taken elsewhere: See, for example, Lawrence (1988) 10 Cr App 464; Brewster (1998) 1 Cr APP R 220, at 226, and Johnas (1982) 2 CCC (3d) 490, at 497; and Terizakis (1986) 41 SASR 252 at 256 and also at 252 where Zelling ACJ said:
      “It cannot be too strongly stated that heroin addiction and the need to obtain money in furtherance of it, is not a matter in mitigation in charges of robbery under arms.”
      230 Reference may also be made to Spiero (1979) 22 SASR 543 at 5488-549, where King CJ with whose reasons for judgment Walters and White JJ concurred, said:
      “one feels sympathy for a person who has become entangled in drug addiction but the Court cannot treat an addiction as an excuse, or even a mitigating factor, in relation to serious crime. Those who are addicted to drugs must understand that if they allow their addiction to lead them into serious crime, they must expect to receive the same severe punishment as would be received by others.”
      231 In Halewyn (1984) 12 A Crim R 202, Young CJ, in whose judgment Kaye and Beach JJ concurred, said (at 203):
      “The offence thus committed was a very serious one which, as I have said, is all too prevalent in this community. It is an offence which the courts are bound to deal with by imposing substantial sentences, and it has been said a number of times in this Court that the fact that an armed robbery is committed to support a drug habit is of little consequence in mitigating the sentence to be imposed.”
      232 In Talbot (1992) 34 FCR 100, Jenkinson J, with whose reasons O’Loughlin and Higgins JJ agreed, said at 105-106:
      “What in my opinion justifies the denial to the addict of leniency in recognition of the relative strength of his temptation to obtain his drug by armed robbery are on the one hand the magnitude of the community’s need of protection from that offence and on the other hand the moral and legal fault which originated the addiction. It seems unfashionable - and, having regard to recently acquired knowledge of molecular biology, perhaps imprudent - to express moral judgments in sentencing criminals. But evaluation of moral culpability remains in my opinion as fundamental to our system of punishment for crime as evaluation of the degree of harm caused or threatened by the criminal act or actor. So much I think the reasons of Mason CJ, Brennan, Dawson and Toohey JJ for their judgment in Veen v The Queen (No 2) (1988) CLR 465 make clear, particularly in their quotation (at 473-474) from an article by C S Lewis (Even more worthy of judicial citation is Lewis’ The Abolition of Man (2nd ed 1946).”
      233 Their Honours did, however, in this case note the special circumstance where the addiction of the offender originated in the administration of a potentially addictive substance in the course of medical treatment, to which his or her consent was, at best, merely formal. The moral culpability of such an offender was said to be different from that of the offender whose addiction originated in the voluntary administering of a substance that he knew (or I would add, should have known) to be addictive, and the use of which is proscribed by the criminal law.
      234 In Hammond (1997) 2 Qd R 195, the Court said at 199-200:
      “The true relevance of drug addiction as a factor contributing to the commission of crime and its effect in the sentencing process has never been adequately explained. In our view it is a factor that may help an offender to the extent of showing that his or her descent into the crime in question was a secondary consequence of desperation produced by a human weakness rather than a primary choice. It may be inferred in many cases (and the present case is very much in point) that the offender would almost certainly never have become a thief but for loss of control of an ordered life through drug addiction. The proposition that the results of drug addiction are self inflicted is half true and therefore dangerous. The offender must of course be held responsible for his or her own actions. The drug addiction is not an excuse; but it is a factor that may tell the court that the real weakness of character is that of a drug addict rather than that of a robber. That may be by no means inconsequential. It is however a two edged factor; it may also tell the court that rehabilitation is going to be difficult.
          Just how these factors are to be applied to particular cases has not been spelled out, and it may be undesirable to attempt to do so. In our view the fact that drug addiction has led to criminal activity can properly lead to a wider understanding of the offender’s character and motivation for the crime. Depending on the circumstances of the particular case this may sometimes show the offender to be less deserving of condemnation than the primary facts, without more, might suggest.”
      235 Later, the Court said at 211:
      “Addiction to drugs at the time of the offence is not an excuse, but may tell the court that the primary weakness of character is that of an addict rather than that of a robber. Such a factor tends to be mentioned along with an unfortunate family life as an explanation of the offender’s descent into bad habits. When the original problem was not something for which the offender was responsible, it is natural to feel some sympathy for the offender though this must not obscure the offender’s direct responsibility for the ultimate criminal act he or she chose to do.”
      236 These observations were relied upon by the respondents as providing some support for the proposition that the fact of drug addiction may properly lead to the culpability of the offender being regarded as less deserving of condemnation than would otherwise be the case. However, it is clear that the Court in Hammond did not question the widely accepted principle that drug addiction is not an excuse, and emphasised that the relevance of drug dependence will depend upon the circumstances of the particular case. The decision does not stand as authority for the general proposition advanced.
      237 Nolan (1998) VSCA 135 also provides only limited assistance to the respondents. Although Buchanan JA referred to drug addiction as a “mitigating” circumstance, the other member of the majority, Butt JA spoke of it as an “explanation”. It was an extreme case, where the impulsiveness of the offence, rather than the mere fact of drug addiction, took the matter outside the general rule recognised in Halewyn.
      238 Douglas (1995) 56 FCR 465 similarly was an extreme case. Von Doussa, Higgins and Nicholson JJ there accepted that, while it is not a mitigating factor that a person commits a crime to feed a drug addiction, nevertheless there will be extreme cases where that principle will not apply. The moral and legal fault for the acquisition of the addiction can properly be considered, for example where the offender becomes addicted at an age when the “degree of judgment” open to him or her is limited.
      239 The submission advanced essentially turned upon the proposition that, within the context of the national concern that currently exists in relation to the scourge of drug dependency, and its effects not only upon individual users and their families, but also upon the community as a whole, some greater flexibility should be exercised in sentencing those whose criminality is substantially linked to an addiction.
      240 In support of this submission reference was made to the views expressed by some academics and researchers that imprisonment provides no deterrent for this class of offender, and may tend only to exacerbate the problem.
      241 Additionally, reference was made to the need to look for new solutions or creative responses for what has proved to be an intractable pattern. Illustrations cited as either having been adopted or under review in this area include the Drug Court, expanded access to rehabilitation programmes, heroin trials, safe injecting rooms, and new forms of chemical therapy in the place of, or as an alternative to methadone (naltrexone, acamprosate, buprenorphine, slow release oral morphine and so on.)
      242 Within that context it was submitted that the bare statement that drug addiction is “not an excuse,” or is “not a factor in mitigation” reflects an approach by the courts which is too rigid. On the contrary, it was submitted a more constructive approach should now be adopted which gives greater recognition to the cravings of drug dependent offender who is forced through the absence of suitable alternatives, to resort to crime to feed a habit.
      243 Reference to the observation of Mildren J in Lewfatt (1993) 66 A Crim R 451, concerning the imprisonment of the typical offender with which the guideline judgment deals, underlies, at least in part this submission. His Honour there said:
      “The prisoner was young, and had undergone significant changes in her outlook. Her prior convictions were drug and alcohol related. She was now determined to stay off drugs and alcohol, and to pursue her ambitions for a career, and accept responsibility as a parent. Her history strongly suggested that she had gone through a difficult period of immaturity, and was now ready to rehabilitate herself. In those circumstances a period of actual imprisonment could well be counterproductive to this process. If one of the main purposes of punishment is to protect society, society’s interests are best served by a sentencing disposition which promotes the rehabilitation of the prisoner, rather than a disposition which may have the opposite effect.”
      244 To this statement can be added the observation of the majority of the High Court in Veen No.2 (1988) 164 CLR 465 at 476:
      The purposes of criminal punishment are various: protection of society, deterrence of the offender and of others who might be tempted to offend, retribution and reform. The purposes overlap and none of them can be considered in isolation from the others when determining what is an appropriate sentence in a particular case. They are guideposts to the appropriate sentence but sometimes they point in different directions.”
      245 In advancing the proposition that drug dependence should be recognised as a significant factor of mitigation, justifying a lesser sentence than might otherwise be appropriate, the respondents sought to draw upon material questioning the extent to which such dependence is a matter of full and free choice. An affidavit was tendered, for this purpose, sworn by Dr. Jurd, a psychiatrist specialising in the treatment of drug offenders, in which he deposed that:
      “drug addiction has a well documented neurobiological or physiological basis.”
      246 In addition he said that the:
      “predisposition to addiction has a substantial physiological (genetic) component. Thirty years of study of the neurobiological basis of alcoholism has revealed time and again that having a positive family history is among the most powerful predeterminants of alcoholism:
          (a) identical twins are more likely than fraternal twins to both suffer from alcoholism.
          (b) adopted away children of alcoholic parents, even if they have never lived with their parents, are more likely to themselves become alcoholic than other adoptees.
          (c) the risk of developing alcoholism increases depending on how many alcoholic relations one has, whether or not you live with them.
          (d) a series of molecular biological experiments have shown that at least one gene is over-represented among alcohol dependent people.
          There is every reason to extrapolate this information to other drugs. Similar studies have been done with drug dependents producing very similar results.”
      247 He continued by noting that drug addiction affects the addicts’ behaviour in predictable ways, altering values and overwhelming other priorities as the addiction becomes more severe.
      248 The hypothesis that addiction has a genetic component or predisposition was not tested. Nor was the validity of the comparison with alcohol dependency. Whether this view be correct or not, I am not persuaded that there is an inevitable causal relationship or progression from drug addiction to the commission of criminal offences, or to an incremental increase in the seriousness of these offences.
      249 In many instances, as studies tend to show, that may well be he case: Dobinson et al Drugs and Crime: A Survey of New South Wales Property Offenders 1984. New South Wales Bureau of Crime Statistics and Research, 56, 57, 62 and 64; Dobinson Making Sense of the Heroin-Crime link (1882) 22 ANZJ Crim, 269; Maher et al Running the Risks: Heroin, Health and Harm in South West Sydney National Drug & Alcohol Research Centre University of New South Wales Monograph No 38, 51 and 62; Hall, Methadone Maintenance Treatment as a Crime Control Measure Bureau of Crime Statistics and Research (1996) 2; Hall The Role of Legal Coercion in the Treatment of Offenders with Alcohol and Heroin Problems (1997) 30 ANZJ Crim 105; Lipton The Effectiveness of Treatment for Drug Abusers under Criminal Justice Supervision (1995) National Institute of Justice, 4.
      250 However, many offenders commence their criminal careers before turning to drugs, and not all persons who become addicted to drugs commit armed robberies or even serious property offences.
      251 Moreover, there are many factors other than drug dependency, such as multiple social disadvantage or an attachment to a criminal sub culture, that contribute to criminal behaviour; and it would, accordingly, be too simplistic to generalise that drug dependency is inevitably the precursor to or cause of criminality or of recidivism.
      252 Nor am I persuaded of the appropriateness of the suggested analogy between drug addiction and mental abnormality, in respect of which the element of general deterrence is often given less weight: Veen (No.2) at 476-477, Scognamiglio (1991) 56 A Crim R 81; Tsaris (1996) 1 VR 398; Letteri (Court of Criminal Appeal New South Wales 18 March 1993) and Engert (1995) 84 A Crim R 67.
      253 The relevant principle as stated in Letteri by Badgery-Parker J, in a passage adopted by Gleeson CJ in Engert is as follows:
      “… that whereas general deterrence is a relevant consideration in every sentencing exercise, it is a consideration to which less weight should be given in the case of an offender suffering from a mental disorder or severe intellectual handicap. In an extreme case, the proper application of this principle may produce the result that considerations of general deterrence are totally outweighed by other factors. In every case it is a matter of balancing the relevant factors in a manner no different from that which is involved in every sentencing exercise.”
      254 The reason for this approach lies in the circumstance that the community will readily understand that the offender who suffers from a mental disorder or abnormality is less in control of his or her cognitive facilities or emotional restraints, and in some instances lacks the ability to make reasoned or ordered judgments. Almost invariably there is a limited appreciation of the wrongfulness of the act, or of its moral culpability, which although falling short of avoiding criminal responsibility does justify special consideration upon sentencing. Moreover, such a condition is inherent and its presence does not depend upon any element of choice.
      255 The case of the drug dependent offender is very different in so far as the community is likely to see such a person as a perfectly appropriate medium to stand as an example to others.
      256 These differences were implicitly recognised in Terizakis (1986) 41 SASR 252 where O’Loughlin J, with whose reasons Cox J agreed, said at 256:
      “… heroin addiction cannot, of itself, justify a sentencing judge reducing a penalty, which is otherwise appropriate to the offence that has been committed and to the circumstances in which the offence was committed. One cannot help but have immense sympathy for a person who is addicted to heroin; and it would be callous to adopt an attitude that, because it is self-inflicted, the addict is not worthy of sympathy or concern. But having said that, it unfortunately remains a fact of life that all too many armed hold-ups are these days committed by persons who, in one way or another, are dependent upon drugs and are seeking money to meet the costs of that dependency. Despite the cravings caused by the addition and despite irrationality that is so often a consequence of the addiction, the majority of drug addicted persons still retain, in varying measures, the ability to distinguish between right and wrong and the ability to comprehend that the consequence of armed robbery will be a lengthy term of imprisonment. To give way, out of sympathy and concern, by imposing lesser terms of imprisonment than would otherwise be imposed, might appear to be an act of humanity, but it will do nothing to stamp out the drug trade and, worse, it might even give some indirect encouragement to would-be offenders.”
      257 The comparison is also imperfect in so far as it overlooks the original element of choice every person has, initially whether or not to experiment with drugs, and thereafter to continue with their use. Even if some persons have a genetic predisposition or vulnerability to addiction, they are not, in my view, entitled to claim any favours in that regard. The addictive quality of drugs and the potential of a slide into a degraded or criminal lifestyle are so well known as to invite a free choice. Those who choose to use drugs must, in my view, accept the consequences.
      258 Moreover, as Gleeson CJ observed in at 71:
      “…the question of the relationship, if any, between the mental disorder and the commission of the offence, goes to circumstances of the individual case to be taken into account in the application of the relevant principles. The existence of such a causal relationship in a particular case does not automatically produce the result that the offender will receive a lesser sentence, any more than the absence of such a causal connection produces the automatic result that an offender will not receive a lesser sentence in a particular case. For example, the existence of a causal connection between the mental disorder and the offence might reduce the importance of general deterrence, and increase the importance of particular deterrence or of the need to protect the public. By the same token, there may be a case in which there is an absence of connection between the mental disorder and the commission of the offence for which a person is being sentenced, but the mental disorder may be very important to considerations of rehabilitation, or the need for treatment outside the prison system.”
      259 By the same process of reasoning the existence of a causal relationship between a drug habit and the commission of an offence should not, as a matter of general principle, automatically lead to a lesser offence.
      260 The respondents also relied upon the proposition that there was little evidence that the punishment of offenders has any general deterrent effect, whether for offenders at large or drug dependent persons in particular. It may be recognised that while there are different views about this, some studies have been interpreted as showing that the perceived severity of a sentence, as distinct from the certainty of detection and arrest, does not of itself provide a deterrent effect: Anderson et al “Formal and Informal Sanctions: A comparison of deterrent effects” (1997) Social Problems 25: 103-117; Paternoster et al “Perceived Risk and Social Control: Do Sanctions really deter?” (1983) Law and Society Review 17: 467-479; Baron et al “Deterrence and Homeless Street Youths (1998) Canadian Journal of Criminology 27-60; James Austin et al “Does Imprisonment Reduce Crime? A Critique of ‘Voodoo’ Criminology” (1993) National Council on Crime and Delinquency. 10, 22.
      261 In the case of drug dependent persons contemplating the commission of an armed robbery, in particular, it was suggested, that the prospect of imprisonment provides little deterrent, in that:
      · their addiction causes them to give high priority to drug seeking behaviour, sometimes to the exclusion of all else;
      · they are accustomed to engaging in high risk activity: Maher et al “Running the Risks: Heroin, Health and Harm in South West Sydney” National Drug and Alcohol Research Centre University of NSW NDARC Monograph no. 38 at 26, l20, 125; Hall Methadone Maintenance Treatment as a Crime Control Measure (1996) Bureau of Crime Statistics and Research, 2.
      · the clearance rate for armed robbery is less than 1 in 5 (Chilvers: New South Wales Recorded Crime Statistics 1998 at 30-32;
      · there is no realistic prospect of law enforcement activity reducing the size of the drug addicted population or of it reducing the street level price of drugs (thereby alleviating the financial pressure to resort to crime to support a habit) Maher supra at 126) and
      · despite the steady increases in sentences, following the introduction of the Sentencing Act 1989, and also amendment of the Crimes Act 1900 to increase the maximum available penalty, in respect of many offences, the crime rate has not fallen.
      262 It was additionally submitted, there was little evidence that lengthy periods of imprisonment have any substantial rehabilitative consequences for the drug dependent offender. To the contrary, the argument ran, and as recidivism rates suggest, incarceration may be counterproductive, initiating or encouraging the offender into criminal behaviour, and breaking social and familial bonds which militate against such behaviour.
      263 Alternative sentencing options, it was submitted, were more likely to reduce drug dependency and the consequent incidence of criminality and of recidivism: Hall: “Methadone Maintenance Treatment as a Crime Control Measure” (June 1996) Contemporary Issues in Crime and Justice, Bureau of Crime Statistics and Research 3, and 6-7; Hall: “The Role of Legal Coercion in the Treatment of Offenders with Alcohol and Heroin Problems” (1997) Vol 30 ANZJ Crim 113; Lipton: “The Effectiveness of Treatment for Drug Abusers Under Criminal Justice Supervision” (1995) National Institute of Justice 51-53; Murphy: “Drug Courts: An Effective Efficient Weapon in the War on Drugs” (1997) Illinois Bar Journal Vol 85 487; Tauber: Drug Courts: a Judicial Manual California Centre for Judicial Education and Research (1994) 1-2 and 9-10; and Belenko “Research on Drug Courts: A Critical Review” (1998) National Drug Court Institute Review Vol 1 Issue 1, 21-23, 29 and 35.
      264 The views expressed in these studies concerning the deterrent value of imprisonment, and the possible exposure of prison inmates to influences that may increase their prospects of recidivism, are not confined to drug dependent offenders. They are of common application, particularly in the case of first offenders, for whom the slamming prison door principle remains apposite, and also in the case of those who are yet to offend, for whom the prospect of imprisonment has a particular relevance.
      265 I am not prepared to advocate any departure from the long accepted wisdom that imprisonment does have a personal and general deterrent effect. It is a notion deeply entrenched in the criminal law, and it has the imprimatur of the legislature which has prescribed significant maximum penalties for the offence under consideration, as well as for other offences involving serious criminality.
      266 Moreover, it cannot necessarily be assumed from the fact that increases in sentences have not been accompanied by any noticeable drop in crime rates, that they lack deterrent effect. In the absence of any control, it cannot be known whether that crime rates would have been higher had sentences not been increased.
      267 The risk of exposure to undesirable influences within a corrective environment, similarly cannot be used as a justification for abandoning, or even limiting imprisonment, as a general response to criminal conduct. There remains for every offender a choice between reform and recidivism, and the problem is better addressed by the development of adequate programs and rehabilitation options within the prison environment, than it is by a significant change in sentencing policy.
      268 While it may be accepted that the rehabilitative aspect of sentencing is not an unimportant consideration: Osenkowski (1982) 30 SASR 21; Douglas (Court of Criminal Appeal New South Wales 4 March 1997); Eastway (Court of Criminal Appeal New South Wales 19 May 1992); Fabian (1993) 64 A Crim R 365; and Halewyn, I am not persuaded that the offender, whose criminality was directly influenced by drug addiction, should be placed into a special category for sentencing purposes.
      269 The objective of rehabilitation is but one aspect of sentencing. It needs to be taken into account along with the objectives of retribution and deterrence.
      270 It is more appropriate, in my view, to have regard to these considerations, upon an individual case basis, in which it is recognised that the offence of armed robbery remains a very serious crime, which, save in exceptional circumstances, calls for full time imprisonment. That is not to say that, in an individual case, the fact of drug dependency of the person standing for sentence may not remain a relevant fact reflecting on the objective circumstances of the offence, and/or the subjective circumstances of the offender, and particularly on the relativity of the minimum and additional terms: Bugmy (1990) 169 CLR 525 at 537; and B (1993) 68 A Crim R 547.
      271 This approach accords with the caution offered by Gleeson CJ in Engert at 68:
      “It is … erroneous in principle to approach the law of sentencing as though automatic consequences follow from the presence or absence of particular factual circumstances. In every case, what is called for is the making of a discretionary decision in the light of the circumstances of the individual case, and in the light of the purposes to be served by the sentencing exercise.”
      272 It also avoids the problems associated with establishing categories according to the offender’s need for money (e.g. to feed a drug addiction, to support a gambling habit, to pay off business or family debts, to establish a business, etc) which would lend themselves to abuse and to moral judgments in respect of which minds may legitimately differ.
      273 In my view the relevant principles are as follows:
          (a) the need to acquire funds to support a drug habit, even a severe habit, is not an excuse to commit an armed robbery or any similar offence, and of itself is not a matter of mitigation;
          (b) however the fact that an offence is motivated by such a need may be taken into account as a factor relevant to the objective criminality of the offence in so far as it may throw light on matters such as:
              (i) the impulsivity of the offence and the extent of any planning for it; (cf Bouchard (1996) 84 A Crim R 499 at 501-2); and Nolan (1988) VSCA 135 (2 December 1998);
              (ii) the existence or non existence of any alternative reason that may have operated in aggravation of the offence, eg that it was motivated to fund some other serious criminal venture or to support a campaign of terrorism;
              (iii) the state of mind or capacity of the offender to exercise judgment, eg if he or she was in the grips of an extreme state of withdrawal of the kind that may have led to a frank disorder of thought processes or to the act being other than a willed act;
          (c) It may also be relevant as a subjective circumstance, in so far as the origin or extent of the addiction, and any attempts to overcome it, might:
          (i) impact upon the prospects of recidivism/rehabilitation, in which respect it may on occasions prove to be a two-edged sword (eg Lewis Court of Criminal Appeal New South Wales 1 July 1992);
          (ii) suggest that the addiction was not a matter of personal choice but was attributable to some other event for which the offender was not primarily responsible, for example where it arose as the result of the medical prescription of potentially addictive drugs following injury, illness, or surgery (cf Hodge Court of Criminal Appeal New South Wales 2 November 1993; and Talbot); or where it occurred at a very young age, or in a person whose mental or intellectual capacity was impaired, so that their ability to exercise appropriate judgment or choice was incomplete;
          (iii) justify special consideration in the case of offenders judged to be at the “cross roads”: Osenkowski(19882) 5 A Crim R 394.
      274 To go further, and to accept the fact of drug addiction as a mitigating factor generally, would not be justified in principle. Moreover, it would involve an exercise in irresponsibility on the part of the Court, if it were understood as a message that committing the crime of armed robbery to feed a drug habit is less deserving of censure than would otherwise be the case.
      275 The legislature has, by the heavy maximum penalty prescribed for armed robbery, spoken clearly in relation to this offence. Drug dependent persons should not be encouraged, as a class, to think that they are free to engage in serious criminal conduct of whatever kind with impunity, or with any hope of favourable treatment because they are able to show that they needed money through their addiction.
      276 In summary, I see no reason to depart from the planks of punishment, retribution and rehabilitation that underlie the sentencing process generally, and that permit of individualised sentencing by reference to the objective and subjective circumstances of each case.
      277 I do not suggest by these observations that greater attention and resources should not be dedicated to drug rehabilitation generally, both within and outside the prison environment. Clearly it is desirable that alternative approaches to the problem be trialed, it being far preferable that potential criminality be headed off than punished after the event.
      **********

      IN THE COURT OF

      CRIMINAL APPEAL
                          60559/98
                          60558/98
                          60561/98
                          60746/98
                          60596/98
                          60595/98
                          60511/98
                                      SPIGELMAN CJ
                                      WOOD CJ at CL
                                      NEWMAN J
                                      HULME J
                                      SIMPSON J
                                  Wednesday 12 May 1999

                REGINA v Paul Anthony HENRY
                REGINA v Stephen Anthony BARBER
                REGINA v Hoai Vinh TRAN
                REGINA v Troy David SILVER
                REGINA v Theo TSOUKATOS
                REGINA v Bill KYROGLOU
                REGINA v John David Jenkins

        JUDGMENT

        278 NEWMAN J: I have read the reasons of Spigelman CJ relating to the implementation of guide lines in respect of sentencing in relation to the crime of armed robbery and whether drug addiction can be taken into account by way of mitigation. I agree with those reasons. I also agree with Wood CJ at CL.
        **********

        IN THE COURT OF
        CRIMINAL APPEAL
                            60559/98
                            60558/98
                            60561/98
                            60746/98
                            60596/98
                            60595/98
                            60511/98
          SPIGELMAN CJ
                                      WOOD CJ AT CL
                                      NEWMAN J
                                      HULME J
          SIMPSON J
          Wednesday 12 May 1999
          REGINA v Paul Anthony HENRY
          REGINA v Stephen Anthony BARBER
          REGINA v Hoai Vinh TRAN
          REGINA v Troy David SILVER
          REGINA v Theo TSOUKATOS
          REGINA v Bill KYROGLOU
          REGINA v John David Jenkins

          GUIDELINE JUDGMENT

          HULME J:
          Guideline Judgment
          279 I have read in draft the Reasons of Spigelman CJ. I agree with his Honour’s conclusion that a guideline judgement is required and generally with His Honour’s reasons for that conclusion. However, on that topic there are some additional remarks I would make.
          280 Firstly, I am satisfied that the incidence of armed robbery has had an impact on society by changing patterns of behaviour out of a sense of apprehension. The increasing use over recent, perhaps 10 or so, years of time delay locks, video cameras and of the practice that safes can only be opened by employees of companies such as Brambles with their armoured vans, I have no doubt, is a reaction to this type of conduct.
          281 Secondly, the evidence adduced by the Crown satisfies me that there has been a substantial increase in the incidence of armed robbery. In this regard four sources of data were presented to the court: Victim Surveys, Court Statistics, Statistics kept by the NSW Bureau of Crime Statistics and Research, and information in the Report on Government Services 1999 released by the Steering Committee for the Review of Commonwealth/State Service Provision.
          Victim Survey Data
          282 The Victim Survey data was contained in annexure B14 to the Crown’s original submissions. This was a May 1998 “Crime and Justice” bulletin published by the NSW Bureau of Crime Statistics and Research. The article in that bulletin deals with “robbery” generically and records that it is based on what was said to be “a representative sample of about 6,500 NSW households involving some 13,000 individuals”. It notes however that there was some “confusion about the definition of robbery (resulting) in a significant over estimate of the incidence of robbery in NSW between 1990 and 1995”. A change in the definition of robbery was then used in the survey with the result that data collected in 1996 and 1997 is not comparable with that collected earlier.
          283 The article purports to record victimisation rates for robbery between 1990 and 1997. The figures for 1996 and 1997 are recorded as 0.4% and 0.3% respectively. The figures for other years have to be derived from a rather small chart and, although precision is impossible, seem to be as set out below. The direction of movement from year to year is a correct reflection of the chart.
          1990 1.7
              1991 1.6
              1992 1.1
              1993 1.3
              1994 1.2
              1995 1.8
          284 The article also recorded that in 1997 over 65% of the respondents to the survey who had been a victim of a robbery indicated that there was no weapon used and less than 3% of the robbery victims indicated that a firearm had been used.
          Criminal Court Statistics
          285 These were extracted from the relevant pages of seven annual publications produced by the NSW Bureau of Crime Statistics and Research and were:-


          YearNumber of Persons ChargedNumber of Charges Finalised
          19918531306
          19929701481
          19938341251
          19947681051
          1995681943
          1996708976
          19977841128
          286
          In its submissions the Crown said that “the publication records the number of people charged separated by the principal offence committed by each person (i.e. the most serious offence) and the total number of charges finalised”. However, reference to page 54 of the full 1996 publication suggests that this is not so and that, whatever other counts a person may have faced, so long as one count was robbery that person is included within the “number of persons charged”. I have not found a clear explanation of the “number of charges finalised”. It is not apparent whether it reflects the number of counts of robbery dealt with or the number of counts dealt with in the case of offenders who faced one count of robbery.
          NSW Bureau of Crime Statistics and Research
          287 The third set of statistics are from publications of the NSW Bureau of Crime Statistics and Research recording information supplied by the police. The Court was informed that methods of recording and correlation have changed over the years and that it is not possible to directly compare figures prepared under one system with those prepared under another. Although one cannot directly compare the figures in one of the periods referred to below with the figures in another, it is possible to compare figures within a period with other figures in the same period.
          1982/83 - 1989/1990
          288 Annexure B11 to the Crown’s original submissions was an extract from the NSW Recorded Crime Statistics 1989. It shows the following rate of recorded offences per 100,000 population.
          Offence82/8383/8484/8585/8686/8787/8888/8989/90
          Robbery with firearm32.522.120.315.516.417.816.514.4
          Robbery with a weapon not a firearm12.09.812.515.416.719.519.722
          Total rows 1 & 244.631.932.830.933.137.236.236.4
          Robbery - no weapon37.637.436.030.137.338.741.644.2
          Total rows 3 & 482.269.368.861.070.475.977.880.6

          Note: The total of rows 3 and 4 has been inserted by me.

          289 So far as armed robbery is concerned, the figure for 1982/83 is so high it looks to be an aberration. The figures show an increase of about 13% between 1983/84 and 1989/90 or between the three year period 83/84 - 85/86 on the one hand and the three year period 87/88 - 89/90 on the other. The figures for “robbery - no weapon” are suggestive - I put the matter no higher - of an increasing trend.
          1991-1993
          290 The information for 1991/93 is much briefer. It comes from the Bureau publication NSW Recorded Crime Statistics 1993 and is contained in Annexures B9-B10 to the Crown’s original submissions. The commentary records that for robbery overall “there was no statistically significant upward or downward trend over the 3 year period to December 1993”. However, it is recorded that “there was a statistically significant downward trend in robbery with a firearm … a 30.7% decrease”.
          April 1994-December 1995
          291 The information covering the period April 1994 to December 1995 is again brief. The Court was informed that it came from an August 1996 Crime and Justice Bulletin published by the Bureau. The extract forms Annexure B8 to the Crown’s original submissions. The commentary records:-
          “From the first 9 months to the last 9 months of the period April 1994 to December 1995, robbery without a weapon increased by 7.3% and robbery with a weapon other than a firearm went up by 27.3%. The recent trend for robbery with a firearm has been neither upward or downward.”
          292 It is not possible from the evidence to say whether this information relates to the absolute number of offences or is related to population. However, one can be confident that the population of New South Wales did not increase at the rate of 27.3% in this period.
          1995-1998
          293 Information for 1995-97 comes from the Bureau publication NSW Recorded Crime Statistics 1997. It is contained inter alia in Annexures B4-B6 of the Crown’s original submissions. This is extended to 1998 in an extract from the 1998 publication which formed Annexure C to the further Crown submissions. Apart from a decrease between 1997 to 1998 in the case of robbery with a firearm and an (insignificant) decrease in the case of robbery without a weapon in the same period, these figures show a substantial increase in offences. The figures per 100,000 population are:-
          Offence1995199619971998
          Robbery with firearm11.113.317.5/18.214.4
          Robbery with weapon not firearm2430.553.4/53.969.2
          Total rows 1 & 235.143.870.9/72.183.6
          Robbery without a weapon73.979.8102.2/103.4101.9
          Total rows 3 & 4109123.6173.1/175.5185.5

          Notes: Totals have again been inserted by me.
              Annexures B4-B6 on the one hand and Annexure C on the other state different figures for 1997.
          294 At face value these figures indicate that between 1995 on the one hand and 1997 and 1998 on the other there has been an increase of over 100% in the incidence of armed robbery, more so where the “arm” has not been a firearm.
          295 Some documents show the absolute number of offences committed during some of the periods referred to above. Annexure B12 to the original Crown submissions and Annexures B4-B6 and C to the further Crown submissions reveal:-

          Offence82/8383/841995199619971998
          Robbery no weapon20042009392643275633/56925511
          Robbery with firearm17331187597732 962/1001 821
          Robbery with weapon not firearm 641 529128916562951/29783861

          Note: Annexures B4-B6 on the one hand and Annexure C on the other state different figures for 1997.

          296 I do not forget the remarks made above about the difficulty of comparing figures for one period with figures in another . However, these figures show that the number of offences of robbery without a weapon has at all times been a substantial proportion of the total, and during 1995-1998 significantly exceeded the total of the number of offences involving a weapon whether or not a firearm. It follows from these figures that the Victim Survey Data and Criminal Court statistics which refer to robberies generically are of no assistance in making any judgment on the question of whether there has been a change in the incidence of armed robbery.
          Report on Government Services 1999
          297 This document is Annexure A to the Crown’s supplementary submissions. It records that on the basis of crimes reported to police the incidence of armed and unarmed robbery per 100,000 of population in Australia and its various states and territories during 1993 to 1997 was as follows:-

          Armed Robbery
          YearNSWVICQLDWASATASACTNTAUST
          199338.622.230.328.333.59.618.18.330.2
          199436.717.327.034.331.813.321.38.228.3
          199562.316.826.638.724.812.925.79.836.7
          199648.618.627.254.823.110.131.814.334.1
          199779.325.736.658.925.39.536.212.248.6
          Unarmed Robbery
          YearNSWVICQLDWASATASACTNTAUST
          199361.520.532.029.782.912.920.419.542.3
          199484.819.033.537.371.318.021.922.850.0
          199594.521.132.942.274.413.527.033.454.5
          199693.623.436.542.367.420.035.145.155.3
          1997121.027.835.059.357.422.037.426.766.1


          298 The combined figures for New South Wales, per 100,000 population are set out in the table below. For the purposes of comparison, I have set out in a third column the comparable Bureau of Crime Statistics figures where available.
          1993 101.1
              1994 121.5
              1995 156.8 109
              1996 142.2 123.6
              1997 200.3 173.1/175.5
          299 Albeit there are some “dips”, the figures for New South Wales show a decided upward trend. Except for South Australia and, in the case of armed robbery, Tasmania, the other states and territories also show an upward trend in both classifications. Australia wide, including New South Wales, there was an increase in each category between 1993 and 1997 of over 50%.
          Conclusion
          300 The inconsistency between the above sets of figures raises the question whether any can be relied upon. At first blush each has doubt thrown on it by one or more of the others.
          301 The change in definition used in the compilation of the Victim Surveys does not inspire confidence. The limited extent of the sample leads to some doubt as to its validity. The proportion of unarmed robberies recorded in that survey of over 65% is not substantially different from the proportion of these recorded in the Bureau of Statistics figures for 1997 (59%), but the proportion of robberies with a firearm - less than 3% - is radically different from the Bureau of Statistics figures over many years. The Bureau information for the periods 1982/83 - 1989/90, 1991 - 1993, and 1995 - 1998 has consistently shown such offences to be a decreasing proportion of robberies as a group but the lowest proportions ever have been in 1997 (10.1 - 10.3%) and 1998 (7.7%). Given the consistency in these Bureau figures and the period involved, the proper conclusion to draw in respect of the survey figures relating to robberies with a firearm is that they are unreliable. That throws considerable doubt on the validity of the survey.
          302 The only possible support any of the other data provide for the Survey figures is in the Criminal Court statistics which, for the period 1992 to 1995 echo to some extent the decreasing trend in the Survey figures for all robberies for the period 1990 to 1994, some time lag in the former figures being expected.
          303 In that the Criminal Court statistics do not take into account offences not leading to a person being charged and it is not clear that they even detail the number of offences the subject of a charge, they have obvious limitations. Neither do they indicate the extent to which at the beginning and end of the periods under consideration, there may have been a backlog of cases. The Criminal Court statistics are also in absolute numbers and I confess to some surprise at the suggestion implicit in those figures that robbery has actually decreased since 1991 or 1992, despite the notorious increase in drug taking which has occurred. On the other hand, the information contained in the Report on Government Services does show decreases in some states over some periods. An example is Victoria between 1993 and 1996.
          304 The Bureau of Crime Statistics information does have a deal of consistency about it over an extended period. In saying that I do not of course suggest that there have not been departures from trend lines and again I do not forget that direct comparison of data between periods is not possible. I have referred to the relative decrease in robberies with a firearm over three of the four periods covered by this information. There is a similar consistency, albeit in the other direction, in the data relating to robbery with a weapon other than a firearm within the periods 1982/83 - 1989/90 (except for 1982/83 itself), April 1994 - December 1995, and 1995 - 1998 and in the data relating to robbery without a weapon within the periods 1982/83 - 1989/90 (except for the first 3 of these 8 periods), April 1994 - December 1995, and 1995 - 1998. In that the information for 1991 - 1993 records that there was no statistically significant trend overall but a significant downtrend in robbery with a firearm, during that period, it indicates that there was an increase in robbery without a weapon or robbery with a weapon other than a firearm, or both, during the period. Given the length of time between 1982/83 and 1998, the exceptions I have noted do not invalidate the conclusions I have drawn as to the consistency of the information.
          305 The difference between the information from the Bureau of Crime Statistics and that in the Report of Government Services suggests that it could not have come from the same source. There are dips in 1994 and 1996 in the armed robbery figures and these are, of course 2 out of the 5 years referred to. On the other hand the figure for 1996 is appreciably higher than for 1994 and the other three years all show an increase. The unarmed robbery figures all tend in the one direction apart from those for 1996 which are only marginally less than for 1995. Considered overall, it seems to me that the information in the Report tends in the same direction as that from the Bureau, albeit for the much shorter period of 1993 to 1997. Thus although the Report figures cast doubt on those from the Bureau, both sets of data support the conclusion that there has been a significant increase in the incidence of offending in New South Wales.
          306 Absent other evidence, I would have been disposed to regard the Bureau of Crime Statistics as a reliable source of information in respect to statistics falling within its province such as those set out above. Having considered them, the Victim Survey data and Criminal Court statistics to which I have referred do not cause me to doubt the reliability of the Bureau statistics. If I had to choose between them, I would be disposed to regard the statistics from the Bureau as more likely to be reliable than that from the Report on Government Services but because of the limited purpose to which it is urged the statistical data should be put, I do not need to make that choice. The information from both the Bureau and the Report demonstrate that there has been a very substantial increase in the incidence of both robbery and armed robbery. The statistics from these two sources are sufficiently consistent and the periods covered by them sufficiently long that, subject to any statutory or discretionary restraints, the Court should approach the issues before it on that basis.
          307 At the risk of repetition, the information from the Report indicates that in New South Wales between 1993 and 1997 there was an increase in armed robbery incidents per 100,000 population from 38.6 to 79.3 - 105%. If the average of 1996 and 1997 is compared with the average of 1993 and 1994, the increase is 70%. The increase in unarmed robbery and robbery overall is about 100%.
          308 The information from the Bureau of Crime Statistics indicates-
          Between 1983/84 and 1989/90, or between the three year period 83/84 - 85/86 on the one hand and the three year period 87/88 - 89/90 on the other, the incidence of armed robbery increased by about 13%.
              In the period 1991 to 1993, in robbery overall there was no statistically significant trend albeit robbery with a firearm decreased by about 30.7%.
              From the first to the last 9 months of the period April 1994 to December 1995, robbery with a weapon other than a firearm went up by 27.3%. The trend for robbery with a firearm was neither upward or downward.
              Between 1995 and 1998 there was an increase in armed robbery incidents per 100,000 population from 35.1 to 83.6 - 138%. The percentage increase is 188% if robbery with a weapon not a firearm is the category considered.
          309 The size of the increase between 1995 and 1998 is staggering. Given that the period covered by the figures for those years is only four years and the change is so much greater than that apparent for earlier periods, I am disinclined to rely on that data as showing reliably in quantitative terms, a long term trend. However, when regard is had to the fact that the New South Wales statistics for three of the four periods mentioned show a substantial increase in the incidence of armed robbery during those periods, and the Government Services report shows the same for the period it covers, both for New South Wales and Australia generally, I am of the view that one should accept that the offence has become far more common in this State, not merely in absolute terms but also per head of population. It must be remembered that insofar as the percentage increases to which reference has been made are accepted, they need to be compounded if one is considering the change since about 1983.
          310 I would add that, given the increasing usage of heroin and other drugs in the community and the need of most addicts to fund their addiction by crime, the conclusion that there has been a substantial increase in the incidence of armed robbery is not surprising.
          311 It was not suggested on behalf of the respondents to the appeals that resort to statistical material of the above kind was other than legitimate and no other inconsistent data was advanced. Certainly, the reliability of the data was criticised but I have dealt with that topic. The Court’s attention was directed to the victimisation rates for 1996 and 1997 of 0.4 and 0.3% respectively and it was submitted that these showed that armed robbery was a reasonably rare offence. However these figures indicate that the average chance of being robbed during a 50 year adult life span is between 15 and 20%. It is not obvious that that prospect should be regarded with equanimity.
          312 The increased incidence in armed robbery, particularly where the weapon is not a firearm, justifies a review of the level of sentencing that has applied hitherto and thus a guideline judgment.
          Evidence as to the Impact on Victims
          313 Turning to another topic, I take the view that the evidence as to the impact of robbery on victims generally is a matter to which the Court may have regard either as a matter of judicial notice or as an “other matter” within s29(1) of the Criminal Procedure Act and a matter the Court is entitled to take into account on the question of sentence in the individual cases. It is of the same nature as the material the Court was referring to in R v Broxam (unreported, CCA, 3 April 1986):-
          “The prevalence of the offence in respect of which the applicant was sentenced is a matter of grave social concern… It is now, well recognised that the victims of such armed robberies or attempted armed robberies may suffer devastating psychological damage consequent upon the trauma necessarily involved with the commission of such offences. In these circumstances sentencing judges would be recreant to the trust which the community places in them, if they were not to impose sentences consistent with the seriousness and prevalence of such offences.”
          314 The passage was cited with approval in R v Bell (unreported, CCA, 11/5/93) and R v Stefanovski (unreported, CCA, 9 June 1994).
          315 That material is relevant to sentencing standards generally and therefore to the individual cases here. Of course, any evidence as to the impact on the particular victims is also relevant whether that impact be negligible or major. But if there is no particular evidence, it does not seem to me that the Court must proceed on the basis that there has been no impact or that it is not entitled to inform itself as to the possible impact.
          316 The full terms of Section 29A of the Criminal Procedure Act 1986 are set forth in paragraph 56 of the judgment of the Chief Justice. Sub-section (1) allows the Court to take into account “evidence or other matters”. Sub-section (2) provides that the “Court must not increase a sentence in any appeal by reason of, or in consideration of, any evidence that is used by the Court in giving a guideline judgment in the appeal and that was not given at the trial”. Normal principles of statutory construction lead to the conclusion that there is no prohibition in sub-section (2) in using “other matters” to increase a sentence. Within this expression would be the material referred to in Veen v R (No 2) (1987-88) 164 CLR 465 at 473 and that to which this Court had regard in R v Hollocoglu (1991) 29 NSWLR 67.
          The Guideline for New South Wales
          317 On the question of what the guideline for New South Wales should be, I agree with Spigelman CJ that the category of case to which His Honour referred is sufficiently common for the purposes of setting a guideline. However, in my view the guideline should be set at a level higher than that proposed by the Chief Justice. It should be set so as to include a total term of penal servitude of 5 years and, except in unusual circumstances, a minimum term of 3 years even when special circumstances within Section 5 of the Sentencing Act 1989 (NSW) are found. Absent special circumstances, the minimum term would be, as required by that section, not less than 3¾ years.
          318 A minimum term should be included in the guideline because inadequacy in the length of such terms is one of the deficiencies in past sentencing practice which is apparent in the review of the Judicial Commission sentencing statistics referred to in the judgment of the Chief Justice. Given the frequency with which special circumstances have been found - and there is nothing to suggest this is likely to change - and minimum terms of less than half the full term have been imposed - matters referred to in paragraph 107 of the Chief Justice’s reasons - it is likely that, absent the inclusion of a minimum term in the guideline, there will continue to be many sentences imposed for armed robbery where the total term is 4 years and the minimum term is 2 years or less. As guidelines, I regard the former figure as too low and the latter as wholly inadequate for a term which must “remain such as appropriately reflects the criminality involved” - Morrissey (unreported, CCA, 15 July 1994, cited with approval in R v McDonald (unreported, CCA, 12 October 1998).
          319 The statutory provisions relevant for present purposes are those contained in Sections 94 to 98 of the Crimes Act 1900 (NSW). Although actions other than robbery with or without aggravating features are encompassed within these Sections, it will be convenient in the discussion which follows to refer to the term rob and its derivatives as encompassing the all of the primary activities proscribed. The sections reveal a clear pattern of seriousness in the mind of the legislature.
          · Simple robbery carries a maximum punishment of penal servitude for 14 years - Section 94.
          · Robbery in circumstances of aggravation, or in company, or being armed with an offensive weapon, or accompanied by the malicious infliction of actual bodily harm carries a maximum punishment of penal servitude for 20 years - Sections 95 and 97(1).
          · Robbery accompanied by:-
              (a)Wounding - Section 96, 98,
                  (b) The infliction of grievous bodily harm - Section 96, 98, or
                  (c) Being armed with a dangerous weapon - Section 97(2),
                  increases the maximum penalty to one of 25 years.
          320 (The inclusion by s4 of a dangerous weapon within the definition of “offensive “ weapon, if the definition applies throughout these provisions, does complicate the orderly pattern otherwise revealed. However, it does not affect the substance of the above picture.)
          321 Commonly the objective circumstances which may be involved in offences against this group of sections may be considered as falling within four areas, viz. the amount of money or the value of the items involved, the premeditation and degree of planning on the part of an offender, the offender’s actions towards his victim or victims and fourthly the impact of those actions on the victim or victims. It is appropriate that, to discourage such activity, the law impose high sanctions on those who participate in criminality for high stakes. Those whose criminality extends over a greater period and displays deliberateness and planning can, all other things being equal, anticipate the duration of their criminal intent and actions to be reflected in any sentence imposed.
          322 In considering the actions of an offender, it should be recognised that robbery is a denial of one of the elementary freedoms on which our society is based. Robbery involves the imposition of the will of the offender by force or the threat of force on the rights and ability of the victim to go about his own affairs in his own way. Some actions on the part of offenders, e.g. being in company, being armed, the malicious infliction of actual bodily harm, find express recognition in the statement of the offences. Others are also relevant. Thus conduct calculated to cause terror rather than merely the fear necessarily involved in an offence is an aggravating circumstance. Relevant also are the risks created by an offender - Readman 47 A Crim R 181 at 185, Lane (unreported, CCA, 3 November 1995). A consideration of the cases demonstrates that not all victims meekly acquiesce in a robber’s demands. Commonly a robbery occurs in circumstances charged with emotion and experience within and without the courts shows that there is but a fine line between a threat and a weapon on the one hand and injury or death on the other. Although an offender who crosses that line may lay him or herself open to further charges, it is important that the courts impose real sanctions to discourage the creation of the risk.
          323 Some impact on the victim is specifically referred to in the statutory provisions. But it is also important to bear in mind that it is an essential element in every robbery that the victim will be put in fear of an offender and of what that offender might do. After all it is the engendering of that fear which it is the offender’s intention and by which he is able to carry out his subjection of the victim or his will.
          324 In many cases the fear will be of imminent death with the concomitant loss of all that life holds and everything the victim holds dear. Experience of life and experience within the courts demonstrates that not all persons readily recover from such traumatic events. In other cases the fear may not be extreme and in not all cases will it continue to have an impact or significant impact on the victim after the threat causing it is removed. Not all people react to circumstances of stress in the same way. However, as I indicated above when referring to R v Broxam, R v Bell, and R v Stefanovski, the potential devastating psychological damage consequent on armed robberies is something of which the Courts have taken judicial notice over many years.
          325 By definition, the category of offence to which the guideline in this case is directed falls towards the bottom end of the scales of amount and premeditation. That is not to suggest these aspects are of no significance. The loss of $1,000 to a small shopkeeper may be more serious to him than the loss of $1M to a bank. The premeditation will rarely be of less than some hours duration and, even if the selection of the victim has not occurred long before the commission of an offence, the intention to commit, and preparation for, the offence will often be considerably longer than this.
          326 The category of offence does not fall so low when the other aspects to which I have referred are considered. The imposition on a shopkeeper is no different in kind from the imposition on a bank. The imposition on, fear engendered in, and the emotional or psychological reaction of the person threatened does not depend on whether they are a shopkeeper, bank teller or jewellery store owner. The risks of harm may be less with a knife than a gun, though this is not necessarily so and both can be fatal. Much will depend on often unpredictable actions of the victim and reactions of an offender who may be desperate for money and the drugs he can obtain with it. Indeed resistance by an uninsured shopkeeper faced with the loss of his own monies may be more likely than opposition from trained staff in larger organisations though one must recognise that in the latter situation there may be more persons exposed to danger.
          327 The commencing point in any consideration of an appropriate sentence for a statutory offence is the statute itself. As was said in another context, “In determining the proper penalty … the fundamental consideration is rather the degree by which, having regard to the maximum penalties provided by the Act in question, the Respondent’s conduct would offend against the legislative objective of suppressing the illicit traffic in the prohibited drug” - R v Peel (1971) 1 NSWLR 247 at 262.
          328 Having regard to the terms of the legislation to which I have referred and the nature of armed robbery, an offence which comes within a “worst case” category for which 20 years is the maximum penalty will commonly involve a large amount of money, a deal of premeditation, imposition on, and the risk of harm to, a number of people and serious psychological harm to at least one. It is likely it will involve “company” and/or an offensive weapon. The offender will commonly, though not necessarily, be a repeat offender. If one is considering a worst case for which 25 years is the maximum penalty, the offence is likely to involve also a dangerous weapon and the injury may be grievous bodily harm. Clearly, the category of case for which the guideline is intended falls a long way short of these sorts of offences. But so do the penalties I propose fall a long way short of 20 or 25 years. Although I do not suggest that anything shorter is insignificant, in absolute terms 5 years is but one quarter and 3 years is but 15% of the 20 year maximum. Terms of 3 and 5 years are shorter than the time spent by many people in tertiary or other training or in jobs which are of no long term significance to them. The periods are relatively short when compared with an adult life span. I said in R v Spiteri [1999] NSWCCA 3 that I regard 10 years imprisonment as more than twice as severe as a 5 years sentence The longer period is likely to have an impact on an offender’s life in terms of wife, children, job prospects and the like from which he may well never recover. Those observations apply a fortiori to a sentence of 20 or 25 years.
          329 In my view nothing less than the terms I proposed is appropriate as a norm or guideline for a premeditated offence of the nature of that described by the Chief Justice and which can be expected to place the victim, and possibly other members of the public, both in fear and at risk of serious injury or death and which is calculated, whatever the offender may intend, to cause significant psychological or other problems in at least part of the population. However small the amount taken, however short may have been the premeditation in a particular case, and however well an individual victim may have coped with the imposition on him or her, society has a great interest in protecting its members from the incidents necessarily attendant upon any armed robbery and in reducing the risks attendant on repetition of armed robberies, even small armed robberies, in the future. The punishment imposed must be such as to amount to a substantial deterrent.
          330 The statistics referred to when I was dealing with the topic of whether there should be a guideline judgment indicate, as I have said, that the offence of armed robbery has become far more common. The correct approach of the Court in that situation is set forth in the judgement of King CJ (with whom the other members of the court agreed) in R v Dube (1987) 46 SASR 118, quoted by the Chief Justice at paragraph 151 of his Reasons, viz. to increase the level of punishment. The increased incidence of the offence argues strongly that the guideline I propose is indeed insufficiently severe. However, as that guideline is a significant increase on the level of sentencing which has prevailed in the past, I am not persuaded that the guideline should be set at a level higher than I propose until one sees the result. After all, a significant number of offenders will receive a higher sentence than I have indicated because of the presence of aggravating circumstances such as a bad prior record and the commission of an offence while on conditional liberty.
          Drug Addiction
          331 I turn then to the topic of the significance of drug addiction in the determination of an appropriate penalty. On this topic I agree with the Reasons of the Chief Justice and Wood CJ at CL. In so doing, I should expressly acknowledge that I retreat from the remarks I made in R v Murray (unreported, CCA, 7 November 1995), quoted by Wood J. I am persuaded both by the reasoning of their Honours and by the authority quoted by them that, at least as a general proposition, what I said in R v Murray was wrong.

          IN THE COURT OF
          CRIMINAL APPEAL

          60559/98
          60558/98
          60561/98
          60746/98
          60596/98
          60595/98
          60511/98

          SPIGELMAN CJ
          WOOD CJ at CL
          NEWMAN J
          HULME J
          SIMPSON J

                                  Wednesday 12 May 1999

          REGINA v Paul Anthony HENRY
          REGINA v Stephen Anthony BARBER
          REGINA v Hoai Vinh TRAN
          REGINA v Troy David SILVER
          REGINA v Theo TSOUKATOS
          REGINA v Bill KYROGLOU
          REGINA v John David JENKINS

          GUIDELINE JUDGMENT


          SIMPSON J :
          332 I have read in draft the judgments of Spigelman CJ and Wood CJ at CL. I agree, for the reasons given by the Chief Justice, that it is appropriate that this court promulgate a guideline in relation to sentences imposed for offences against s 97 of the Crimes Act, and I agree with the guideline sentence proposed at para 165 of the draft judgment.
          333 I would, however, with respect to the offence/offender profile outlined in paragraph 162 of the draft judgment, express a preference for the promulgation of a guideline that relates to sentencing after conviction following a trial. (This would, of course, require lengthier terms than those specified in the proposed range.) There are three reasons for my preference. The first is that not all pleas of guilty carry the same weight, depending as they do on the strength of the Crown case, and the time at which they are offered. If the starting point assumed a plea of not guilty, then the appropriate discount to reflect the true worth of the plea could be given. The second reason is that the severity of the sentence to be imposed in relation to any offence might be mitigated by numerous facts or features, of which a plea of guilty is only one. I see no reason to single out the plea of guilty from other mitigating features in the definition of the kind of offence to which the guideline applies. The third and most substantial reason concerns appearances. Although the truth is that the guideline sentence has built into it a discount representing the benefit to which an offender is entitled for pleading guilty, the appearance, when an offender is sentenced following a trial, will be of a penalty that is attributed to the exercise of the right to the put the Crown to proof. The appearance would, as I have said, be false, but would exist nonetheless.
          334 However, as I understand that other members of the Bench do not share these concerns, I am content to concur in the guideline as proposed.
          335 I wish to say something about the submission advanced on behalf of certain of the respondents to the effect that drug addiction when a causal factor in the commission of offences should be recognised as a significant mitigating feature.
          336 It is a mistake, in my opinion, to regard drug addiction as a starting point. It is an end point, or a point on the way to the end, of a process. Where the process begins may vary in individual cases but it does not necessarily begin when a person decides to use a prohibited substance. It probably has its origins well before the date of the first use of drugs. In the worst, or least forgivable, cases it may have its origins in arrogance, in an antipathetical attitude to the laws of society, or in weakness of character. In other cases, I have no doubt, it has its origins in social disadvantage, poverty, emotional, financial, or social deprivation, poor educational achievement, unemployment, and the despair and loss of self-worth that can result from these circumstances or any combination of them. In this court one sometimes sees cases in which drug taking stems from sexual assault or exploitation, sometimes committed when the person who turns to drugs, and who comes before the court, is very young, and sometimes the precipitating events have occurred many years before. Drug addiction is not always the disease; it is, as often as not, a symptom of social disease.
          337 Drug addicts do not come to their addiction from a social or environmental vacuum. This court should not close its eyes to the multifarious circumstances of disadvantage and deprivation that frequently precede and precipitate a descent into illegal drug use. I do not suggest for a moment that all drug users fall into this category. It is because some do and some do not that I believe rigid rules about the impact on sentencing of drug dependency on sentencing cannot be laid down.
          338 I cannot accept that the blameworthiness of one drug taker is (even excepting that small number of individuals who begin drug taking with medically prescribed drugs) always to be treated as being at the same level as the blameworthiness of the next. Nor can I accept that the exercise of free choice in the use of drugs is always of equal dimensions. It is not every decision to use drugs that can properly or fairly be characterised as a decision made in the exercise of free choice. The will of an individual can be overborne or undermined, not only by acts of another person, but also by the pressure of circumstances. I do not accept that most drug offenders are truly exercising free will when they choose the degradation, despair, criminality and cycle of imprisonment that can follow the initial use of illegal drugs. The circumstances that propel the offender to the use of drugs are often, if not usually, beyond his or her control. They may or may not be combined with a vulnerable personality or even a weakness of character. Many drug offenders have not had the life experiences or the normal developmental path that permit a conclusion that the decision to take drugs was a decision made in the exercise of a free choice in the sense in which that phrase is ordinarily understood.
          339 An analogy can legitimately and properly be drawn with the principles relating to the sentencing of aboriginal offenders, collected and stated by Wood J (as he then was) in R v Fernando (1992) 76 A Crim R 58 at 62 - 63. As I understand the remarks on sentence in that case, his Honour observed that aboriginality alone does not excuse serious crime nor operate as a mitigating factor. However, since it is often associated with other circumstances of disadvantage or deprivation, aboriginality may explain or throw light on the particular offence or the circumstances of the offender. It is, in truth, not aboriginality, but the life experiences too commonly associated with aboriginality that are relevant to the sentencing decision. Just as it is properly said that, while drunkenness is not normally an excuse or mitigating factor, where alcohol abuse reflects the socioeconomic circumstances and the environment in which an offender has grown up, that fact can and should be taken into account as a mitigating factor (Fernando, principle E, p 62), so also can it properly be said that drug abuse may reflect the socioeconomic circumstances and the environment in which another offender has grown up, and that may equally then be taken into account as a mitigating factor.
          340 Wood J went on to recognise the great social difficulties sometimes faced by aboriginal communities where poor self image, absence of education and work opportunity and other demoralising factors have placed heavy stresses on them, reinforcing their resort to alcohol and compounding its worst effects. So, too, it is with some drug offenders. The passage from which the above is drawn can readily be adapted to the situation of some drug offenders. Where great social difficulties, poor self image, absence of education and work opportunity and other demoralising factors have placed heavy stresses upon drug offenders, causing or reinforcing their resort to drugs, then it is appropriate for this court to give the same recognition to those antecedent circumstances.
          341 A sentencing judge always has a delicate balancing task to perform, taking into account the demands of the community for retribution, deterrence both general and specific, and the interests of the same community in the rehabilitation of an offender. The exercise may call for an examination of the circumstances that led the offender to drug use, addiction and crime. All the circumstances that precipitate the use of drugs are relevant to the evaluation of moral culpability that is essential to the sentencing process.
          342 It would, therefore, be too simplistic to lay down a principle that addiction either is, or is not, a mitigating circumstance in the sentencing of offenders convicted of drug related crime. I much prefer the approach encapsulated in paragraphs 270 - 272 inclusive of the draft judgment of Wood CJ at CL, favouring the “individualised justice” to which Mahoney ACJ referred in R v Lattouf (unreported, NSWCCA, 12 December 1996).
          343 While, on this approach, it might be possible to argue that moral culpability for the offence is diminished, suggesting a more lenient rather than a more severe penalty, the counter balance is that, absent evidence of positive prospects of a cure of the drug addiction, a finding that the offence is unlikely to be repeated (itself ordinarily seen as a factor in favour of a lesser sentence) would be unavailable. Absent positive evidence of real rehabilitation, the diminution in moral culpability which results from drug addiction originating in the kind of social or familial deprivation to which I have referred cannot result in a corresponding diminution in sentence. Evidence of rehabilitation is the key to the impact evidence of drug addiction might have in the sentencing process. This view I perceive to be largely in accord with the direction taken by the legislature in the establishment of a Drug Court, specifically created to divert drug offenders with realistic prospects of rehabilitation (other than those involved in violence) from the criminal justice system: Drug Court Act 1998 (NSW) S 5(2)(b); para 4 of the draft judgment of Wood CJ at CL.
          344 Where circumstances such as those I have mentioned (or others equally deserving of compassion) have been the foundation for the drug addiction, and part of the causal chain leading to the commission of crime, then it would be appropriate, in my view, for the rehabilitative aspects of sentencing to assume a more significant role than might otherwise be the case. In an appropriate case, rehabilitation might outweigh other sentencing factors. In order for those circumstances to provide a reason for reduction of sentence, however, there would need to be strong evidence of real progress towards actual rehabilitation. I would not wish to be understood to be saying other than that leniency of the kind to which I refer depends heavily upon demonstrated (as distinct from theoretical) rehabilitative prospects.
          345 The authorities cited by Spigelman CJ and Wood CJ at CL are almost unanimous in holding that drug addiction of itself does not operate as a mitigating circumstance. When the words “of itself” are emphasised, I respectfully agree with the proposition repeatedly stated; but, in any event, the line of authority is so compelling and so sustained that, even sitting as one of a bench of five, I would be reluctant to participate in a decision departing from it. It is the role of the legislature (or the High Court) to alter a such a long standing principle. What I have said above is consistent with that line of authority . Nowhere is it held that it is inappropriate, in sentencing drug offenders, to take into account either the circumstances that gave rise to the drug addiction or demonstrated rehabilitation.
          346 In all the mass of material put before this court to support the argument that drug addiction should be treated as a mitigating factor, there was nothing to inform the court, statistically or otherwise, of what causes drug addiction, or its socioeconomic background. What I have said, therefore, is somewhat impressionistic, much of it drawn from experience on this court, and some of it assumptions based upon more general reading. It is, I believe, largely uncontroversial. It is not intended as a recitation of the most universal or common circumstances pre-dating or causing drug addiction, but of some familiar patterns.
          347 I have not lost sight of the material to which the court was taken concerning the effect of armed robberies on their victims. Obviously, the offence is serious, commonly has drastic consequences for victims, and must be punished accordingly.
          348 I agree that the principles stated by Wood CJ at CL in paragraph 273 of his draft judgment are the principles that emerge from the authorities. To them, I would add that not all drug addicts are necessarily to be taken to have brought themselves to addiction and criminality with the same degree of foresight, determination, informed decision making, and consequent moral culpability. To impute those qualities to every decision to take drugs is to ignore the important starting point, the reasons underlying the initial drug use.
          349 I entirely agree with the proposition that the bare fact that an offence is motivated by a need for money to support a drug habit does not, alone, mitigate the offence or operate to reduce the sentence to be imposed.
          350 Counsel who appeared for some respondents also argued, by analogy with the sentencing of mentally abnormal offenders, that, in sentencing drug addicted offenders, it may be appropriate to afford less weight to the principle of general deterrence. This argument cannot be accepted. In considering general deterrence it is necessary to consider the class or pool of individuals at whom deterrence is directed. In s 97 offences, that class or pool is the class of people who commit, or are tempted to commit, the offence of armed robbery - many of whom, as the materials show, are also motivated by the need for money to feed drug addiction. These are people whose will not to commit crimes is already weakened by the drug dependence. That means an even stronger message must be conveyed. Far from diminishing the importance of general deterrence, the fact that the population in which deterrence is necessary is unlikely to be deterred other than by firm action on the part of the courts speaks for greater rather than lesser emphasis on that principle.
          351 Usually, and certainly in relation to armed robbery, general deterrence is a very significant factor in the sentencing process. So is specific deterrence and so is punishment. The point I wish to make is that where a combination of two circumstances exists, then general deterrence, and other sentencing objectives such as retribution, may yield to rehabilitation. The two circumstances are:
          352 (i) a background to the addiction that may explain the offender’s decision to use drugs, and diminish his or her moral culpability for that decision;
          353 (ii) demonstrated progress towards rehabilitation.

          354 The second is, to my mind, essential before general deterrence can be seen to give way to rehabilitation.
          355 Where those two circumstances coincide, then the interests of the community may well be better served by the imposition of a penalty that leans towards furthering the rehabilitative process at the expense of the punitive and even the deterrent objectives of sentencing.
          356 I do not understand this approach to be significantly different from the recognition given by Wood CJ at CL to the need for sentencing to be sufficiently flexible to take account of all relevant circumstances (paragraphs 270 - 2719). It is consistent with many authorities in this court and other jurisdictions: see, eg R v Lattouf (unreported, NSW CCA, 12 December 1996); R v Hayes (1987) 29 A Crim R 452 at 457, 472; R v Eager (unreported, NSW CCA 1 November 1995); R v Crotty (unreported, NSW CCA 28 February 1994); R v Fabian (1992) 64 A Crim R 365 at 372, 378, 380; R v Molina (1984) 13 A Crim R 76 at 77 (Federal Court of Australia) (quoted with approval in R v Ellis (1993) 68 A Crim R 449 by Kirby P, who, however, dissented in the result in that case); see also the remarks in Molina in relation to the offender there before the court at p 79.2; R v Dowie (1989) 42 A Crim R 234 at 247 (Court of Criminal Appeal, Tasmania); R v Osenkowski (1982) 30 SASR 212 at 212-3 (Supreme Court of South Australia in Banco); R v Halewyn (1984) 12 A Crim R 202 at 205-6 (Court of Criminal Appeal, Victoria).
          **********
          - 7 -
          IN THE COURT OF
          CRIMINAL APPEAL


          60559/98
          60558/98
          60561/98
          60746/98
          60596/98
          60595/98
          60511/98


                                      SPIGELMAN CJ
                                      WOOD CJ at CL
                                      NEWMAN J
                                      HULME J
                                      SIMPSON J


                                      Wednesday 12 May 1999

                      REGINA v Paul Anthony HENRY
                      REGINA v Stephen Anthony BARBER
                      REGINA v Hoai Vinh TRAN
                      REGINA v Troy David SILVER
                      REGINA v Theo TSOUKATOS
                      REGINA v Bill KYROGLOU
                      REGINA v John David JENKINS

          Six Crown appeals and one severity appeal involving the offence of armed robbery under s97(1) (one involving the aggravated offence under s97(2)) of the Crimes Act 1900 (NSW) were heard together. The Crown submitted that it was appropriate to promulgate a guideline judgment with respect to this offence.

          Held (By the Court):
          Sentencing Guidelines: General
          The purpose of a guideline judgment is to foster consistency in sentencing: Jurisic NSWCCA 12 October 1998; Lowe (1994) 154 CLR 606 at 610-611.

          The appropriateness of an appellate court promulgating guidelines has been authoritatively established: Norbis v Norbis (1986) 161 CLR 513.

          A guideline judgment on the subject of sentencing does not lay down anything in the nature of a rule. The failure to sentence in accordance with a guideline is not itself a ground of appeal: Norbis per Brennan J at 536, 537-538; Jurisic at pp15,16.

          Consideration of the Canadian and South Australian approaches to sentencing guideline. McDonnell (1997) 114 CCC (3d) 4336; Police v Cadd (1997) 94 ACrimR 466; Bini (1994) 68 ALJR 859, referred to.

          Use of New Material in Individual Cases
          Consideration of the use of material before the appellate court that was not before the sentencing judge. The word “evidence” in s12 of the Criminal Appeal Act 1912 and s29A of the Criminal Procedure Act 1986 is confined to “matters going to the commission of the offence or the personal circumstances of the accused”: Beldan (1986) 21 ACrimR 159; Chanh Nghia Ly NSWCCA 16 December 1992; J (1992) 64 ACrimR 441 at 459; Veen v The Queen (No 2) (1987-88) 164 CLR 465; Hallocoglu (1991) 29 NSWLR 67.

          Therefore criminal statistics on the incidence of an offence, including police data, court data and victim surveys are not “evidence” within s12 of the Criminal Appeal Act 1912 nor s29A of the Criminal Appeal Act 1986. Whilst academic literature on the effects of an offence including both physical and psychological effects, may not be “evidence” within these sections, (Hulme J dissenting), it is not material to which the Court should have regard in the individual cases as the gravity of each case is determined by, inter alia, the effects on the particular victims.

          Per Hulme J:
          Observations on criminal statistics.

          Need for Guidelines: Armed Robbery
          The increased incidence of a particular offence is a relevant factor when considering whether or not a guideline judgment is appropriate and, if so, what change with respect to past sentencing practice is called for. Whilst armed robbery is a relatively rare crime, its rate of occurrence has significantly increased in recent years.

          Furthermore, the sentencing statistics of the Judicial Commission of New South Wales indicate:
          (i) Non-custodial sentences are not confined to exceptional cases
          (ii) Leniency is suggested in the full terms
          (iii) The pattern of leniency is reinforced by the extent to which the statutory ratio between minimum and additional terms has been varied
          ((iv) The large proportion of lenient sentences suggest inconsistency in sentencing practices.

          The sentencing statistics suggest both inconsistency and systematic leniency, which justify the promulgation of a guideline judgment.

          Sentencing Guidelines: Armed Robbery
          (a) Type of penalty
          Armed robbery is not simply a crime against property. It is a crime against persons. The fear engendered by the perpetrator of this crime, together with the continued adverse effects on its victims, establish armed robbery to be a serious crime which required condign punishment. Non-custodial sentences for this offence should only be imposed in exceptional circumstances: Roberts (1994) 73 ACrimR 306 at 308. The number of cases in which non-custodial sentences have been imposed indicates that this principle has not been implemented by sentencing judges. The sentencing statistics suggest that this approach is also reflected in the length of custodial sentences when imposed. (Waldron NSWCCA 3 March 1994. See also Ellis (1993) 68 ACrimR 449 at 462; Lawson NSWCCA 12 December 1997).

          (b) Guideline precedents
          Guidelines for armed robbery sentencing in English, New Zealand, Hong Kong, Canadian, Western Australian, South Australian and Queensland authorities referred to. (Turner (1975) 61 CrAppR 67; Gould (1983) 5 CrAppR(S) 72; Attorney-General’s References Nos 3, 4, 8, 9, 10, 11 and 16 of 1990 (Dickson & Ors) (1991) 92 CrAppR 166; Attorney-General’s Reference No 14 of 1991 (Reed) (1992) 13 CrAppR(S) 446; Attorney-General’s Reference NO 7 of 1992 (Khan) (1993) CrAppR(S) 122; Moananui (1983) NZLR 537; Mo Kwong Sang (1981) HKLR 610; Brennan and Jensen (1975) 11 NSR (2d) 541; Chaisson (1975) 24 CC (2d) 159; Johnas (1982) 2 CCC (3D) 490; Norman WACCA 1 February 1989; Miles (1997) 17 WAR 518; Spiero (1979) 22 SASR 543; Dube (1987) 456 SASR 118; Fermaner (994) 61 SASR 447; Drumgoon SACCA 20 November 1995; Hammond (1996) 92 ACrimR 450).

          (c) Guidelines
          A category of case which is sufficiently common for the purposes of determining a guideline comprises the following elements:
          (i) Young offender with no or little criminal history
          (ii) Weapon like a knife, capable of killing or inflicting serious injury
          (iii) Limited degree of planning
          (iv) Limited, if any, actual violence but a real threat thereof
          (v) Victim in a vulnerable position such as a shopkeeper or taxi driver
          (vi) Small amount taken
          (vii) Plea of guilty, the significance of which is limited by a strong Crown case.

          Per Simpson J:
          A plea of guilty should not be included in the definition of the kind of offence to which the guideline applies.

          (d) The appropriate range
          A sentencing range is appropriate in relation to this offence because the seven identified characteristics do not exhaust the factors relevant to sentencing, and many of the characteristics contain within themselves an inherent variability.

          Sentences for an offence of this character should generally fall between four and five years for the full term. Aggravating and mitigating factors will justify a sentence below or above the range, which is itself a starting point.

          Per Hulme J:
          The guideline should be set so as to include a total term of penal servitude of 5 years and, except in unusual circumstances a minimum term of 3 years.

          (e) Circumstantial factors
          A number of circumstances are particular to the offence of armed robbery, including: (a) nature of the weapon; (b) vulnerability of the victim; (c) position on a scale of impulsiveness/planning; (d) intensity of threat, or actual use, of force; (e) number of offenders; (f) amount taken; and (g) effect on victim/s.

          Per Spigelman CJ:
          Drug addiction is a circumstance relevant to the sentencing exercise, but it is not itself a mitigating factor. The existence of a causal relationship between drug addiction an the commission of an offence should not automatically result in a lesser sentence. Valentini (1989) 46 ACrimR 23 at 25; Halewyn (1984) 12 ACrimR 202; Nolan [1998][ VSCA 135; Douglas (1995) 56 FCR 465 at 470; Spiero (1979) 22 SASR 543 at 549; Terizakis (1986) 41 SASR 252 at 256; Hammond (1996) 92 ACrimR 450 at 455-456, 467.

          Per Wood CJ at CL:
          Discussion of general principles of deterrence and drug addiction. Drug addiction is not analogous to mental abnormality in respect of which the element of general deterrence may be given less weight. The relevant principles are as follows:
          (a) The need to acquire funds to support a drug habit is not an excuse to commit an armed robbery, and of itself not a matter of mitigation.
          (b) The fact that an offence is motivated by such a need may be taken into account as a factor relevant to the objective criminality of the offence, insofar as it may throw light on matters such as (i) the impulsivity of the offence and the extent of any planning for it; (ii) the existence or non-existence of any alternative reason that may have operated in aggravation of the offence; and (iii) the state of mind or capacity of the offender to exercise judgment.
          (c) It may also be relevant as a subjective circumstance, insofar as the origin or extent of the addiction, and any attempts to overcome it, might (i) impact upon the prospects os recidivism/rehabilitation; (ii) suggest that the addiction was not a matter of personal choice but was attributable to some other event for which the offender was not primarily responsible; and (iii) justify special consideration in the case of offenders to be at the “cross roads”: Osenkowski (1982) 5 ACrimR 394.

          Per Simpson J:
          Discussion of general principles of rehabilitation and drug addiction. Where a combination of the following two circumstances exist, then general deterrence and other sentencing objectives such as retribution, may yield to rehabilitation: (i) a background to the addiction that may explain the offender’s decision to use drugs, and diminish his or her moral culpability for that decision; (ii) demonstrated progress towards rehabilitation.
          - 6 -
          IN THE COURT OF
          CRIMINAL APPEAL


          60559/98
          60558/98
          60561/98
          60746/98
          60596/98
          60595/98
          60511/98


                                      SPIGELMAN CJ
                                      WOOD CJ at CL
                                      NEWMAN J
                                      HULME J
                                      SIMPSON J


                                      Wednesday 12 May 1999

                      REGINA v Paul Anthony HENRY
                      REGINA v Stephen Anthony BARBER
                      REGINA v Hoai Vinh TRAN
                      REGINA v Troy David SILVER
                      REGINA v Theo TSOUKATOS
                      REGINA v Bill KYROGLOU
                      REGINA v John David JENKINS

          JUDGMENT

          1 SPIGELMAN CJ: In Jurisic NSWCCA 12 October 1998, this Court indicated that it would in future be prepared to issue guideline judgments with respect to sentencing for particular offences. The Court has listed and heard together six Crown appeals and one severity appeal, six of which involves the offence of armed robbery under s97(1) of the Crimes Act 1900 (NSW) and one, the aggravated offence under s97(2). The Crown has submitted that it is appropriate to promulgate a guideline judgment with respect to this offence.
          2 In Jurisic, in a judgment with which the other four members of the Court agreed, I said:
          “In my opinion, guideline judgments should now be recognised in New South Wales as having a useful role to play in ensuring that an appropriate balance exists between the broad discretion that must be retained to ensure that justice is done in each individual case, on the one hand, and the desirability of consistency in sentencing and the maintenance of public confidence in sentences actually imposed, and in the judiciary as a whole, on the other.” (14)
              “Such guidelines are intended to be indicative only. They are not intended to be applied to every case as if they were rules binding on sentencing judges. Decisions of appellate courts on sentencing are not to be treated as binding precedents.” (14-15)
          3 In Jurisic I referred to the numerous cases in which this Court has made statements of general principle with a view to guiding sentencing practice in trial courts. The promulgation of formal guideline judgments, labelled as such, was a development of that practice.
          4 In Jurisic, Wood CJ at CL said:
          “The Court has … over the years endeavoured to lay down sentencing principles for particular classes of case where sentences reflecting a significant element of general deterrence are required, or when non-custodial options are inappropriate. It appears that sometimes these principles are lost or that their significance is overlooked, in the volume of appellant decisions handed down and in the pressures imposed on trial courts to dispose of increasingly busy criminal lists.
              By tagging selected decisions as guideline judgments, the Court is not to be taken as usurping the function of the legislature, or as inappropriately intruding into the exercise of the sentencing discretion reserved to trial judges. Rather, what is intended is for the Court of Criminal Appeal to highlight the sentencing principles which fall for it to determine, in a way that might assist trial judges, the DPP and trial counsel, and reduce the occasion for that degree of inconsistency or departure from principle that is an indicator of injustice.” (1-2)
          5 Subject to any relevant statutory requirements, the sentencing task involves the exercise of a broad discretion which, centuries of practical experience strongly indicate, is best conferred on trial judges.
          6 The circumstances in which it is appropriate for appellate courts to interfere with discretionary decisions of this character are confined. In the case of Crown appeals against sentence, even more stringent restrictions have been applied.
          7 The ineluctable core of the sentencing task is the process of balancing overlapping and contradictory objectives. At the appellate level, that characteristic extends to the balancing of the objectives of consistency and individualisation. The Court must sentence both the offender and the offence.
          8 During the course of these proceedings the Respondents to the Crown appeals and the Appellant in the severity appeal, relied on certain observations of Mahoney ACJ in Lattouf (NSWCCA 12 December 1996) where his Honour repeated his own remarks in Kable v DPP (1995) 36 NSWLR 374 at 394:
          “If justice is not individual, it is nothing.”
          9 This ringing phrase must not be taken out of context. In Lattouf his Honour emphasised the multiple objectives served by the sentencing process. One could equally well say “If justice is not consistent, it is nothing”.
          10 As His Honour put it in Lattouf:
          “General sentencing principles must be established, so that the community may know the sentences which will be imposed and so that sentencing judges will know the kind and the order of sentence which it is appropriate that they impose. But, of course, principles are necessarily framed in general terms. General principles must, of their nature, be adjusted to the individual case if justice is to be achieved. For this reason, it is in my opinion important in the public interest that the sentencing process recognise and maintain a residual discretion in the sentencing judge … There is a public interest in the adoption and articulation of sentencing principles which will deter the commission of serious crime and punish those who commit it … But there are other interests to which the sentencing process must have regard; these are other objectives which the sentencing process must seek to achieve. Paramount amongst these is the achievement of justice in the individual case.” (7)
          11 There is no conflict between the system of guideline judgments established by Jurisic and the reasoning of Mahoney ACJ in Lattouf.


          Guidelines: General
          12 As I indicated in Jurisic, the purpose of a guideline judgment is to foster consistency in sentencing. The importance of consistency was well expressed by Mason J in Lowe (1994) 154 CLR 606:
          “Just as consistency in punishment - a reflection of the notion of equal justice - is a fundamental element in any rational and fair system of criminal justice, so inconsistency in punishment, because it is regarded as a badge of unfairness and unequal treatment under the law, is calculated to lead to an erosion of public confidence in the integrity of the administration of justice.” (610-611)
          13 The appropriateness of an appellate court establishing guidelines has been authoritatively established. In Norbis v Norbis (1986) 161 CLR 513, the High Court had before it an issue concerning the power of the Full Court of the Family Court to lay down guidelines with respect to the exercise of statutory discretions by trial judges. By majority, the Court held that the promulgation of such guidelines was permissible.
          14 Mason and Deane JJ said:
          “It has sometimes been said by judges of high authority that a broad discretion left largely unfettered by Parliament cannot be fettered by the judicial enunciation of guidance in the form of binding rules governing the manner in which the discretion is to be exercised. However, it does not follow that, because a discretion is expressed in general terms, Parliament intended that the Court should refrain from developing rules or guidelines affecting its exercise. One very significant strand in the development of the law has been the judicial transformation of discretionary remedies into remedies which are granted or refused according to well settled principle. It has been a development which has promoted consistency in decision making and diminished the risks of arbitrary and capricious adjudication. The proposition referred to in the beginning of this paragraph should not be seen as inhibiting an appellate court from giving guidance, which falls short of constituting a binding rule, as to the manner in which the discretion should be exercised. And despite the generality of some of the statements to which we have referred, there may well be situations in which an appellate court will be justified in giving such guidance the force of a binding rule by treating a failure to observe it as constituting grounds for a finding that the discretion has miscarried.
              The point of preserving the width of the discretion which Parliament has created is that it maximises the possibility of doing justice in every case. But the need for consistency in judicial adjudication, which is the antithesis of arbitrary and capricious decision making, provides an important countervailing consideration supporting the giving of guidance by appellate courts, whether in the form of principles or guidelines. The tension between the two considerations, each of fundamental importance in family law, has inevitably led to a near dilemma for the Full Court of the Family Court. To avoid the risk of inconsistency and arbitrariness which is inherent in the system of relief involving a complex of discretionary assessments and judgments, the Full Court, as a specialist appellate court with unique experience in family law in this country, should give guidance as to the manner in which these assessments and judgments are to be made. Yet guidance must be given in a way that preserves, so far as it is possible to do so, the capacity of the Family Court to do justice according to the needs of the individual case, whatever its complications may be. Reconciliation of these goals suggests that in most, if not all, cases the Full Court of the Family Court should give guidance in the form of guidelines rather than binding principles of law.” (519-520 internal references omitted)
          15 The other member of the majority in Norbis was Brennan J. He agreed with the reasons of Mason and Deane JJ with one exception:
          “The proposition with which I am unable to agree is this: that an appellate court that gives guidance as to the manner in which a statutory discretion should be exercised may prescribe that such guidance should have the force of a binding legal rule. The authority of an appellate court to give guidance is not to be doubted. It is inevitable that the wisdom gained in continually supervising the exercise of a statutory discretion will find expression in judicial guidelines. That is not to invest an appellate court with legislative power but rather to acknowledge that, in the way of the common law, a principle which can be seen to be common to a particular class of case will ultimately find judicial expression. The orderly administration of justice requires that decisions should be consistent one with another and decision making should not be open to the reproach that it is adventitious … An unfettered discretion is a versatile means of doing justice in particular cases, but unevenness in its exercise diminishes confidence in the legal process.” (536)
          16 His Honour went on to say:
          “It is one thing to say that principles may be expressed to guide the exercise of a discretion; it is another thing to say that the principles may harden into legal rules which would confine the discretion more narrowly than the Parliament intended.” (537)
          17 His Honour also said:
          “There may well be situations in which an appellate court will be justified in setting aside a discretionary order if the primary judge, without sufficient grounds, has failed to apply a guideline in the particular case. Where there is nothing to mark the instant case as different from the generality of cases, the failure will suggest that the discretion has not been soundly exercised. The distinction between such a guideline and a binding rule of law, though essential, may be thin in practice. But the distinction must be maintained and a failure to apply the guideline cannot be treated as an error of law: a failure to apply the guideline is no more than a factor which warrants a close scrutiny to the particular exercise of the discretion. What cannot be shut out is the discretion of a primary judge not to apply the guideline when the circumstances of the particular case show that its application would produce an unjust or inequitable result or that another approach would produce a more just and equitable result.
              The only compromise between idiosyncrasy in the exercise of the discretion and an impermissible limitation of the scope of the discretion is to be found in the development of guidelines from which a judge may depart when it is just and equitable to do so - guidelines which are not rules of universal application, but which are generally productive of just and equitable orders. If it is possible to develop such guidelines, it is possible to ensure order and consistency in the exercise of the discretionary jurisdiction under the Family Law Act.” (537-538)
          18 The appropriateness of an appellate court laying down guidelines for the exercise of a statutory discretion has also been noted in a range of cases, invariably with reference to the desirability of consistency in judicial decision making. These cases include:
          · the statutory discretion to award costs in criminal proceedings (Latoudis v Casey (1990) 170 CLR 534 at 541-542, 558-559, 562).
          · the exercise of the discretion to award costs in civil matters (Oshlack v Richmond River Council [1998] HCA 11; 72 ALJR 578; 152 ALR 83 at [35], [134] and [65])
          · the exercise of a discretion to order the winding up of a company (FAI Insurances Limited v Goldleaf Interior Decorators Pty Ltd (No 2) (1988) 14 NSWLR 644 at 646-647, 657-658 and 660-661).
          · the exercise by a licensing magistrate of a discretion to grant a liquor licence (Shreeve v Martin (1969) 72 SR(NSW) 279 at 289-291).
          · the exercise of a statutory discretion to grant leave to commence proceedings out of time (Salido v Nominal Defendant (1993) 32 NSWLR 524 at 532-533, 535-539, 541).
          · The exercise of a discretion to grant leave to appeal from decisions of arbitrators (Pioneer Shipping Limited v BTP Tioxide Limited [1982] AC 724 at 742-743; FAI Insurances v Goldleaf Interior Decorators supra at 661; Leighton Contractors Pty Limited v Kilpatrick Green Pty Limited [1992] 2 VR 505 at 516, 517, 521).
          19 The High Court has confirmed that a Court may establish guidelines for the exercise of an inherent jurisdiction, namely the parens patriae jurisdiction to act in the best interest of the child. (Secretary, Department of Health and Community Services v JWB and SMB (1991) 175 CLR 218 at 260).
          20 The Full Court of the Supreme Court of Victoria has applied the approach to guidelines in Norbis to the determination of guiding principles for the exercise of a discretionary power of the Court in its inherent jurisdiction, namely the power to order the dismissal of proceedings for want of prosecution. (Maysell v Transport Industries Insurance Co Limited [1995] 2 VR 328 at 334-335).
          21 In the case of matters covered by common law principles - like sentencing for New South Wales offences - there is no inhibition which may arise from the prospect of confining a statutory discretion, which the Parliament has conferred in an unconfined form. (Special considerations will arise in formulating guidelines for Commonwealth offences by reason of s16A of the Crimes Act 1914).
          22 As noted above, in Norbis the joint judgment of Mason and Deane JJ contemplated the possibility of a guideline having “the force of a binding rule”, with the consequence that failure by a trial judge to apply a guideline may constitute a ground for finding that the exercise of the discretion miscarried. It was in this respect that Brennan J disagreed.
          23 The Full Court of the Family Court has adopted the option contemplated by Mason and Deane JJ:
          “In our view the time has come to regard a departure from a longstanding guideline … without adequate explanation as a ground for finding that the exercise of discretion has miscarried.” Docters Van Leeuwen (1990) 14 FamLR 130 at 134.
              (See also Lalor (1989) 14 FamLR 282 at 285; Joshua (1997) 22 FamLR 203 at 214; Nygh “Should the Full Court Offer More Guidance to Judges Sitting at First Instance” (1993) 7 Australian Journal of Family Law 137 esp at 146-149).
          24 However, a differently constituted Full Court of the Family Court has stated that guidelines developed by the Full Court should not be regarded as rules of law:
          “The guidelines that we propose are simply guidelines; they are not rigid rules of law and it does not follow that a departure from them will necessarily vitiate a judgment, although judges, judicial registrars and registrars should, we think, give sufficient reasons for departing when they consider a departure is appropriate.” (Re K (1993) 117 FLR 63 at 81).
          25 The balance of authority strongly supports the reasoning of Brennan J in Norbis. See Shreeve v Martin supra at 290 per Walsh JA; FAI Insurances v Goldleaf Interiors supra at 661 per McHugh JA; Latoudis v Casey supra at 558-559 per Dawson J, with whom Brennan J agreed; Leighton Contractors v Kilpatrick Green supra at 516-517 per Fullagar J with whom McGarvie and Ashbee JJ agreed; Maysell v Transport Industries Insurance supra at 334-335 per Brooking, Teague and Hedigan JJ; Oshlack v Richmond River Council supra at [35] per Gaudron and Gummow JJ, and at [134] per Kirby J.
          26 A similar issue has arisen in Canada. The Supreme Court of Canada has unanimously affirmed that it is appropriate for a criminal appellate court to lay down guidelines in the nature of a starting point for sentencing for a particular offence. McDonnell (1997) 114 CCC (3d) 436 at pars 43, 58-61. (See generally Bloos and Renke “Case Comment: Stopping Starting Points R v McDonnell” (1997) 35 Alberta LR 795).
          27 In McDonnell both (majority and minority) judgments rejected the proposition that departure from a previously decided starting point could, itself, constitute an error of principle that would justify appellate intervention ( at pars 32-34; 100, 109).
          28 The majority accepted that departure from a starting point may “suggest” that a sentence is outside the permissible range but is not “determinative” of that proposition (par 43). The minority seemed to regard departure from the starting point as being more strongly indicative of legal error, in terms of manifest inadequacy (or “demonstrably unfit” in the Canadian terminology) (pars 98-101). Neither judgment adopted the approach which Mason and Deane JJ contemplate in Norbis.
          29 A guideline judgment on the subject of sentencing should not lay down a requirement or anything in the nature of a rule. The failure to sentence in accordance with a guideline is not itself a ground of appeal. Guidelines are not rules of universal application. They may be departed from when the justice of a particular case requires such departure. I made this clear in Jurisic when I said:
          “Such guidelines are intended to be indicative only. They are not intended to be applied in every case as if they were rules binding on sentencing judges. Decisions of appellate courts on sentencing are not to be treated as binding precedents.” (15)
          30 In this respect I refer to my adoption in Jurisic (at 15) of the analysis by Dunn LJ in De Havilland (1993) 5 CrAppR(S) 109 at 114, to the effect that decisions on sentencing are not authorities binding on lower courts in the way decisions on substantive law are binding. I went on to say:
          “ … such guidelines are not binding in any formal sense. They represent a relevant indicator, much as trial judges have always regarded statutory maximum penalties as an indicator”. (Jurisic at 16)
          31 Nevertheless, where a guideline is not to be applied by a trial judge, this Court would expect that the reasons for that decision be articulated, so that the public interest in the perception of consistency in sentencing decisions can be served and this Court can be properly informed in the exercise of its appellate jurisdiction.
          Guidelines: Sentencing
          32 In Jurisic I outlined the development of sentencing guideline judgments in England. It is not necessary to repeat that analysis.
          33 In Canada, a starting point approach to sentencing guidelines has been endorsed by the Supreme Court in McDonnell supra. McLachlin J said:
          “The starting point approach to sentencing involves two steps. First, the judge determines the range of sentences for a typical case. Using that range as a starting point, a trial judge then adjusts the sentence upward or downward on the basis of factors relating to the particular offence and offender … the starting point approach combines general considerations relating to the crime committed with personalised considerations relating to the particular offender and the unique circumstances of the assault.” (par 58)
          34 Her Ladyship went on to say:
          “The traditional notion that sentencing is primarily a matter of impression for the sentencing judge and only secondarily a matter of principle began to be questioned by the Courts in the mid 60s. Behind the challenge lay increasing recognition that some measure of uniformity was essential in a sentencing process that was not only just, but was perceived to be just.” (par 65)
              and
          “The starting point approach appears to meet both the requirements of uniformity and consistency in sentencing and individualised justice.” (par 78)
              (See also pars [23-24, 43]; Ruby Sentencing (4th ed, 1994) esp pp481-482).
          35 The judgments in the Supreme Court referred with approval to the decision of the Alberta Court of Appeal in Sandercock (1985) 22 CCC (3d) 79 where the Court said:
          “The sentencing process now adopted by this court is to state typical categories with precision, and to acknowledge at the same time that each actual case presents differences from the archetypical case. These differences might mitigate or aggravate.” (83)
          36 This is similar to the approach adopted by this Court in Jurisic.
          37 The practice of the Full Court of the Supreme Court of South Australia is to promulgate sentencing standards in the form of an “appropriate sentence range”. However, departure from the range is not itself error.
          38 In Police v Cadd (1997) 94 ACrimR 466, Doyle CJ said:
          “It is the function of this Court to ensure that the sentencing process in individual cases is not affected by error of principle. It is also the function of this Court to ensure that sentences are neither excessive nor inadequate. The latter function is performed in two ways. First, in individual cases, by correcting a particular sentence that is considered to be excessive or inadequate. Secondly, by establishing standards of sentencing for particular offences, when the court thinks it appropriate to do so. This may be done over time through the process of correcting individual sentences. But it may also, in my opinion, be done by the court indicating an appropriate sentence range for a particular offence or offences of a particular type.” (479-480)
          39 His Honour went on to cite a number of cases in which the Court had indicated appropriate sentencing ranges and referred, with approval to the following remarks of Cox J in King (1988) 48 SASR 555 at 557-558; 34 ACrimR 412 at 414-415:
          “… this case is about sentencing standards, but it is important, I think, to bear in mind that when a sentence is created, either by cumulative force of individual sentences or by a deliberate act of policy on the part of the Full Court, there is nothing rigid about it. Such standards are general guides to those who have to sentence in the future …”
          40 Doyle CJ went on to say in Cadd:
          “The function of establishing appropriate standards is an important aspect of ensuring, as best one can, that adequate standards of punishment are observed. Establishing appropriate standards also tends to ensure that there is such consistency of approach as is achievable in a system in which the appropriate sentence depends upon, in part, the circumstances of the individual case and of the individual offender, and a system in which sentencing is as individualised as it is in our system.” (480)
              (See also 487 per Duggan J; 488 and 490-491 per Mullighan J; 511 per Lander J; 520 per Bleby J).
          41 I express my agreement with the following observations of Lander J:
          “There are a number of reasons why this Court can and should indicate in a general way that a particular class of offences should be dealt with within certain standards. First it gives general guidance to judges of this Court and to the courts below of what the superior court within the State considers an appropriate range of penalties for a particular class of offence. Secondly an indication by this Court of an appropriate standards is likely to best ensure consistency of penalty throughout the State. Thirdly it is, in my opinion, part of the deterrent process that this Court pronounces, not only for the assistance of the courts below, but for the general education of the public and the particular education of those who may be likely to offend. The public should know in advance that offences of a particular kind will be likely to lead to a penalty in a range indicated by this Court.
              I agree therefore that it is appropriate for this Court to indicate a standard in relation to some offences. However not all offences are capable of being categorised in that manner. Some offences such as assaults and manslaughter depend so very much upon the circumstances of the particular offence that they cannot be considered in a general way as being within any particular sentence range.” (511)
          42 The South Australian practice was affirmed by the High Court when rejecting a special leave application from one of four sentencing standard cases which had been heard together and which I will consider further below. Brennan J said:
          “By their decisions, the Court of Criminal Appeal hoped to ‘remind sentencing judges of the great importance of maintaining adequate standards of punishment in sentencing for armed robbery’. That being the object of the Court of Criminal Appeal, this Court should not grant special leave to review the range of sentences which the Court of Criminal Appeal has set.
              However, the Court of Criminal Appeal is bound to apply general principles of sentencing to any case in which the Crown seeks to have a range of sentencing established or confirmed so that the actual sentence in any case properly reflects its unique circumstances.” (Bini (1994) 68 ALJR 859)
              Statutory Scheme for Robbery
          43 Chapter 1 of Pt 4 of the Crimes Act 1900 (NSW) creates a hierarchy of offences with respect to robbery, an offence distinguished from other forms of stealing by an element of threat or force, putting a person in fear:
          Maximum Penalty 14 years
              (i) Robbery or assault with intent to rob from the person of another (s94).
              Maximum Penalty 20 years
              (ii) The offence in (i), in circumstances of aggravation, which includes use of violence or malicious infliction of actual bodily harm or deprivation of liberty (s95).
              (iii) Robbery or assault with intent to rob being armed with an offensive weapon or in company (s97(1)).
              Maximum Penalty 25 years
              (iv) The offence in (ii), with actual wounding or infliction of bodily harm (s96).
              (v) The offence in (iii), when armed with a dangerous weapon (s97(2)).
              (vi) Wounding or inflicting grievous bodily harm before, at the time of or immediately after a robbery, when armed with an offensive weapon (s98).
          44 In Brown (1989) 17 NSWLR 472 in a joint judgment of Gleeson CJ, Newman and Loveday JJ, this Court said:
          “As is the case in relation to various other subject matters, the provisions of the Crimes Act 1900 dealing with robbery establishes a series of offences, in ascending degrees of seriousness, and with ascending orders of maximum penalty, depending upon the circumstances of the case … It is apparent that within that range of offences there may exist an enormous variety of individual facts and circumstances which will accompany the commission of a particular crime.” (473)
          45 The cases now before the Court are concerned only with s97(1), save for one conviction under s97(2). The section provides:
          “97(1) Whosoever, being armed with an offensive weapon, or instrument, or being in company with another person, robs or assaults with intent to rob, any person, or stops any mail, or vehicle, railway train, or person conveying a mail, with intent to rob, or search the same, shall be liable to penal servitude for 20 years.
              (2) A person is guilty of an offence under this subsection if the person commits an offence under subsection (1) when armed with a dangerous weapon. A person convicted of an offence under this subsection is liable to penal servitude for 25 years.”
          46 Section 4(1) defines “offensive weapon or instrument” to:
          “… include a dangerous weapon and also include an imitation or replica of an offensive weapon or of an offensive weapon, or an instrument as the case may require.”
          47 “Dangerous weapon” means a firearm, a prohibited weapon within the Prohibited Weapons Act 1989 (NSW) or a spear gun.
          48 It is with respect to the offence under s97(1) that the Court was invited by the Crown to promulgate a guideline judgment.
          Crown Material
          49 The Crown sought to put before the Court a range of materials which were not before the sentencing judge in any of the cases. It is material of a general character relevant to sentencing which can be, and sometimes is, put before a sentencing judge. This material was said by the Crown to be relevant to the determination of the individual cases, as well as to the formulation of sentencing guidelines of general application.
          50 The Crown also put before the Court sentencing statistics for s97(1) as compiled by the Judicial Commission of New South Wales. No objection was taken to the Court referring to this material, either in the individual cases or for purposes of the formulation of guidelines. Such information is, of course, available to all sentencing judges and, accordingly, this Court can act on the basis that the trial judge either had, or could have had, access to it if he or she wished. The Court of Criminal Appeal must, of course, be in the same position in this regard. All parties conceded that access to this information was appropriate.
          51 The Respondents to the Crown appeals and the Appellant in the severity appeal objected to the Court taking into account the additional material in the individual cases. This material was of three kinds:
              (i) Crime statistics on the incidence of armed robbery offences, including police data, court data and victim surveys.
              (ii) Academic literature on the effects of armed robbery on victims, including both physical and psychological effects.
              (iii) A report by a clinical psychologist, prepared for purposes of this hearing, on the psychological impact of armed robbery upon victims.
          52 The objection to the use of this evidence for the individual cases was based on the proviso to s12(1) of the Criminal Appeal Act 1912. That section is in the following terms:
          “(1) The court may, if it thinks it necessary or expedient in the interests of justice:
              (a) order the production of any document, exhibit, or other thing connected with the proceedings, and
                  (b) order any persons who would have been compellable witnesses at the trial to attend and be examined before the court, whether they were or were not called at the trial, or order any such persons to be examined before any judge of the court or before any officer of the court, or justice, or other person appointed by the court for the purpose, and admit any deposition so taken as evidence, and
                  (c) receive the evidence, if tendered, of any witness (including the appellant) who is a competent, but not a compellable witness, and
                  (d) where any question arising on the appeal involves prolonged examination of documents or accounts, or any scientific or local investigation, which cannot, in the opinion of the court, be conveniently conducted before the court, the court or any judge thereof may refer the question for inquiry and report to a commissioner appointed by the court, and act upon the report of any such commissioner so far as the court thinks fit, and
                  (e) appoint any person with special expert knowledge to act as assessor to the court in any case in which it appears to the court that such special knowledge is required for the determination of the case;
                  and exercise in relation to the proceedings of the court any other powers which may for the time being be exercised by the Supreme Court on appeals or applications in civil matters, and issue any warrant or other process necessary for enforcing the orders or sentences of the court: Provided that in no case shall any sentence be increased by reason of, or in consideration of any evidence that was not given at the trial. (Emphasis added)
          53 Section 12 was part of the original Criminal Appeal Act when it was enacted in 1912. The Act as at that time did not make any provision for appeals by the Crown. By s6(3) of the Act as originally enacted, this Court was expressly empowered to increase the sentence on an appeal by a convicted person. This necessitated the proviso in s12.
          54 This is the explanation for the proviso, with which I agree, given by McPherson J in Beldan (1986) 21 ACrimR 159 at 167, where the Court was concerned with the equivalent provision in s671B of the Criminal Code (Qld), inserted in 1913. I also agree with his Honour’s conclusion that the proviso to s12(1) applies to Crown appeals, even though they were subsequently introduced.
          55 In Behar NSWCCA 14 October 1998, I made reference to the possible difficulty which s12 may present for the preparation of guideline judgments. Behar was handed down a few days after Jurisic. Subsequently the Parliament amended the Criminal Procedure Act 1986 with the introduction of Pt 8 “Sentencing Guidelines” by the Criminal Procedure Amendment (Sentencing Guidelines) Act 1998. Section 26 of the Act now empowers the Attorney General to apply to the Court of Criminal Appeal for a guideline judgment.
          56 Section 29A now provides:
          “(1) Nothing in section 12 of the Criminal Appeal Act 1912 limits the evidence or other matters that the Court may take into consideration in giving a guideline judgment (whether or not on application under section 26) and the Court may inform itself as it sees fit.
              (2) The Court must not increase a sentence in any appeal by reason of, or in consideration of, any evidence that is used by the Court in giving a guideline judgment in the appeal and that was not given at the trial.”
          57 By reason of the words “whether or not on application under s26” in 29A(1), the material on which the Crown sought to rely, can be taken into account by this Court in preparing a guideline judgment with effect in futuro. This was conceded by all parties.
          58 Section 29A puts beyond doubt, in the context of the promulgation of sentencing guidelines, what would probably have been accepted even without legislative provision. The formulation of guidelines, even with respect to the exercise of statutory discretions, often requires information which goes beyond the circumstances of an individual case.
          59 The recognition that judges do in fact develop and change the law, as distinct from merely declaring the law, has received general acceptance comparatively recently. Accordingly, the circumstances in which the Court will receive material, including evidence in the formal sense, for purposes of such policy development have never been clearly articulated and established. (See Davis “The Judiciary - Maintaining the Balance” in Finn (ed) Essays on Law and Government vol 1 pp283-285; Doyle “Implications of Judicial Law Making” in Saunders (ed) Courts of Final Jurisdiction: The Mason Court in Australia esp at pp96-97). It is plain that cases which raise policy issues cannot be allowed to become law reform commission inquiries.
          60 The means of acquiring information for the purposes of policy development should not be confined by the rules of evidence developed for fact finding with respect to matters that only concern the parties to a particular case. (See Davis “An Approach to Problems of Evidence in the Administrative Process” (1942) 55 HarvLRev 364 esp at 402; Davis “Judicial Notice” (1955) 55 Col. LR 945 esp at 952-953; Carter “Judicial Notice” in Campbell and Woller (eds) Well and Truly Tried (1982) at pp92-94; Ligertwood Australian Evidence (3rd ed, 1998) par 6.42).
          61 Similar issues have arisen with respect to the determination of constitutional facts. As Brennan J has put it:
          “There is a distinction between a judicial finding of a fact in issue between parties upon which a law operates to establish or deny a right or liability and a judicial determination of the validity or scope of a law when its validity or scope turns on a matter of fact. When a court, in ascertaining the validity or scope of a law, considers matters of fact, it is not bound to reach its decision in the same way as it does when it tries an issue of fact between a parties. The validity and scope of a law cannot be made to depend on the course of private litigation.” (Gerhardy v Brown (1984) 159 CLR at 141-142).
          62 His Honour went on to refer to Breen v Sneddon (1961) 106 CLR at 406-406 and Commonwealth Freighters Pty Ltd & Boland v Sneddon (1959) 102 CLR 280 at 292 and added:
          “The court may, of course, invite and receive assistance from the parties to ascertain the statutory facts, but it is free also to inform itself from other sources. Perhaps those sources should be public or authoritative, and perhaps the parties should be at liberty to supplement or controvert any factual material on which the court may propose to rely, but these matters of procedure can wait consideration on another day. The court must ascertain the statutory facts ‘as best it can’ and it is difficult and undesirable to impose an a priori restraint on the performance of that duty.”
          63 This reasoning is applicable to the determination of sentencing principles or the promulgation of a guideline judgment operative in futuro.
          64 Different considerations apply in the sentencing context, when material relevant at a policy level is sought to be relied upon, either directly or indirectly, by the application of the new or differently articulated policy, principle or guideline, to the case on appeal. Issues of double jeopardy and disparity arise in this context.
          65 The Respondents to the Crown appeal, and the Appellant in the severity appeal, rely on the proviso to s12(1) of the Criminal Appeal Act, as reinforced by s29A(2) of the Criminal Procedure Act 1986. These provisions reflect the principle of double jeopardy. The Respondents submit that none of the material should be taken into account with respect to the particular cases before the Court.
          66 No submission was made that either the criminal statistics or the academic literature could not have been taken into account by the sentencing judge. The focus of the submissions was on the double jeopardy element, reflected in the two statutory provisions. Accordingly, it is unnecessary to consider whether Re Richardson [1920] SALR 24, which turns on the “generally known” formulation from Holland v Jones (1917) 23 CLR 149, is still good law in the light of the different formulation in s144 of the Evidence Act 1995.
          67 Nor was any submission made that, in any case, there was an inadequate opportunity to deal with the additional material (cf s144(4) Evidence Act 1995).
          68 The Crown eventually conceded that the report of the clinical psychologist expressly prepared for use in these proceedings, cannot be relied on in the individual cases. This leaves two categories of material:
              (i) Crime statistics
              (ii) Academic literature on victim impact.
          69 The issue turns on the proper construction of the word “evidence” in each of the proviso to s12(1) of the Criminal Appeal Act and s29A(2) of the Criminal Procedure Act. It is plain that in the later Act, the Parliament intended to use the word in the same sense as it was used in the proviso to the former Act.
          70 It is noteworthy that there is a distinction between subs (1) and subs (2) of s29A of the Criminal Procedure Act in this regard. Subsection (1) refers both to “evidence” and “other matters”, whereas subs (2) refers only to “evidence”.
          71 The principal thrust of the submissions for the Respondents to the Crown appeals was that the additional material on which the Crown wished to rely is “evidence” because the Crown seeks to prove as facts the following:
              (i) The existence of an increase in the number of armed robberies in recent times.
              (ii) The extent and intensity of the impact of armed robberies on victims.
          72 Both of these, it is said, are issues of “fact” about which the Crown seeks to adduce “evidence” in the form identified.
          73 Merely to identify an issue as one of “fact” does not lead to the conclusion that the Court is restricted to “evidence” with respect to its determination. For example, the meaning of an ordinary English word has long been held to be a matter of fact not law, yet the Court consults dictionaries and other books in order to determine the meaning. (See Australian Gaslight Company v The Valuer General (1940) 40 SR (NSW) 126 at 137).
          74 An appellate court may take judicial notice on appeal of something which has not been “noticed” at first instance. (See Warren v Pilkington (1960) Tas SR 6 at 9-15). This must, of course, be subject to any contrary statutory provisions.
          75 The Court may “notice” certain notorious facts of which ordinary persons are presumed to be aware. The Court can “notice” such facts either directly, or after being “reminded” of such facts by appropriate information or evidence. (See eg Holland v Jones (1917) 23 CLR 149 at 153). The Court may make reference to works of reference or authority in order to “remind it” of what it “knows”. (See McQuaker v Goddard (1941) KB 687 at 700-701). Sometimes evidence will be required.
          76 The information which the Court acquires in this way is not “evidence strictly so called”. (See McQuaker v Goddard supra at 700; Baldwin & Francis Limited v Patents Appeal Tribunal [1959] AC 663 at 691; Saul v Menon (1982) NSWLR 314 at 325; Wigmore on Evidence (3rd ed) par 2568a; Stone & Wells Evidence: Its History and Policies (1991) p174; Noakes “The Limits of Judicial Notice” (1958) 74 LQR 59 at 63). Nor, in my opinion, is “knowledge”, of which proof is not required under s144 of the Evidence Act, “evidence strictly so called”.
          77 The proviso to s12, and its interstate equivalents have been the subject of judicial consideration. In Beldan supra, the Queensland Supreme Court held that the proviso did not preclude the receipt by the appellate court of evidence to show that the sentencing process was affected by deception on behalf of the Respondents to a Crown appeal (Beldan supra at 167 and 178). This Court has applied Beldan and admitted evidence to show that the sentencing judge was deceived by evidence of a promise to give assistance to the Crown which, this Court found, the Respondent had no intention to honour. (See Chanh Nghia Ly NSWCCA 16 December 1992 at 10-11. See also Todhunter v Attorney-General (Com) (1994) 52 FCR 228 at 246).
          78 The issue also arose in J (1992) 64 ACrimR 441, where Olsson J, with whom King CJ and Mulligan J agreed, held that the South Australian equivalent section (s359 of the Criminal Law Consolidation Act 1925 (SA)) did not prevent the receipt of evidence as to the conduct of the Respondent to a Crown appeal in failing to honour an undertaking to provide assistance to the Crown in other proceedings. His Honour said:
          “There is no doubt that the clear intention of the Parliament was to ensure that, so far as possible, an offender was not unfairly placed in double jeopardy; and that, where further evidence not given at trial is admitted on appeal, that ought not, in the normal course, lead to an increase in the sentence imposed.

              When the proviso to s359 speaks of evidence that was not given at the trial it is, in my opinion, directing its attention to matters going to either the commission of the offence or the personal circumstances of the accused. It is not concerning itself with the situation in which the Crown is merely informing the court that the person sentenced has failed to honour a specific undertaking given to the court as an express basis upon which the sentence is imposed.” (459)
          79 In Veen v The Queen (No 2) (1987-88) 164 CLR 465, one of the issues before the Court was whether the intermediate Court of Appeal should have received material about the state of psychiatric services in New South Wales gaols as facts relevant to the determination of sentencing principles. In the event the Court held that it was unnecessary to do so but added, by way of obiter dictum:
          “It is unnecessary to consider the procedure adopted by the Court of Criminal Appeal in admitting evidence as to the state of gaol psychiatric services but, subject to appropriate procedural safeguards, we would not regard that court as being precluded from receiving further material relevant to the adoption of sentencing principles or sentencing policy of general application.” (473)
          80 Notwithstanding the reference to “policy of general application”, the issue in Veen (No 2) concerned the use of any such “policy” in the particular case before the court.
          81 This passage from Veen (No 2) was applied by this Court in Hallocoglu (1991) 29 NSWLR 67, where the Court received a document, over objection, being the booklet supplied to prisoners serving periodic detention outlining the operations of that scheme. The Court expressly rejected the submission that such material constituted “evidence” within the meaning of s12(1) of the Criminal Appeal Act. Hunt CJ at CL said:
          “That submission must be rejected … We are entitled to inform ourselves from such material as to the true consequences of an order that a sentence be served by way of periodic detention.” (73-74)
          82 It may be that when Hunt CJ at CL held in Hallocoglu that s12 did not prevent the Court informing itself about the administration of the periodic detention system, he had in mind a test for “evidence” similar to that propounded by Olsson J in J: “matters going to the commission of the offence or the personal circumstances of the accused”. Whilst Hallocoglu was, in certain respects, a guideline judgment, the use of this material was not restricted to the formulation of sentencing principles. The Court had reference to the material for the purposes of the individual case.
          83 I am unable to distinguish the criminal statistics on which the Crown seeks to rely in this case, from the material relied upon by the Court in Hallocoglu.
          84 In my opinion, it should now be accepted that “evidence” in s12 of the Criminal Appeal Act and s29A of the Criminal Procedure Act is confined to “matters going to the commission of the offence or the personal circumstances of the accused”. On this basis the criminal statistics are not “evidence” within the proviso.
          85 The academic literature on the effects on victims is in a different position. Each individual case had its own victim or victims. The objective gravity of each case is determined by, inter alia, the effects on victims in the particular case. The general literature is only of assistance at the level of sentencing principle. Whilst it may not be “evidence” within the proviso to s12, it is not material to which the Court should have regard in the individual cases.
          Crime Statistics and Need for Guidelines
          86 The Crown sought to rely on crime statistics for the purpose of indicating that sentencing guidelines were required by reason of the prevalence of the offence. Prevalence is a relevant consideration when deciding an appropriate level of sentence. (See eg Giles v Barnes (1969) SASR 174 at 181; Peterson (1983) 11 ACrimR 164 at 167-168, 169; Henderson (Jana Brian) (1991) 58 ACrimR 369 at 373; Cuthbert (1967) 86 WN (pt 1) NSW 272 at 278; Colman (Anthony Keith) NSWCCA 5 November 1997; Kukunoski NSWCCA 17 August 1989; Retton (Timothy Craig) NSWCCA 14 November 1994). Prevalence has been acknowledged as a reason for establishing a guideline in English guideline judgments. (See eg Cunningham (1993) 96 CrAppR 422 at 425; Brewster (1998) 1 CrAppR 220 at 224 per Lord Bingham).
          87 “Prevalence” may refer to a situation in which a particular crime occurs with such frequency that it has a salience beyond the persons immediately affected by the crime and, accordingly, impacts on society by changing patterns of behaviour out of a sense of apprehension. None of the material presented by the Crown to this Court suggests that armed robbery is “prevalent” in this sense. Indeed the Crown submissions referred to armed robbery as “a relatively rare crime”.
          88 Rather, what the Crown sought to prove was that the incidence of this particular crime had increased over recent years, so that the objective of general deterrence was entitled to greater weight than it might have hitherto received.
          89 Three sources of data were presented to the Court: police statistics, court statistics and victim surveys. By reason of changes in the manner in which recorded crime statistics were kept by the NSW Bureau of Crime Statistics and Research over the years, police statistics do not permit long term trends to be analysed. The data presented to this Court was broken up into four periods: 1982-1989, 1991-1993, 1994-1995 and 1995-1997. Of the statistics for these respective periods none, save the last three year period between 1995-1997, suggests any changes of statistical significance.
          90 The Crown relied on the fact that for the period 1995-1997 there was a significant increase in the number of armed robberies involving a weapon other than a firearm. The rate per 100,000 for such offences increased from 30.5 in 1996 to 53.9 in 1997 and reached 69.2 in 1998. (NSW Bureau of Crime Statistics and Research NSW Recorded Crime Statistics 1998 p41). It is unlikely that, in the case of this offence, such an increase could have been caused by an increased propensity to report the commission of offences.
          91 The second source of data was the NSW Criminal Court statistics which covered the period 1991-1997, during which there does not appear to be any significant change. However these figures do show fluctuations over the years, both up and down.
          92 The third source of statistics relied upon by the Crown was victim surveys. This information indicated that over the period 1990-1997 there was no evidence of any systematic upward or downward trend with respect to robbery victimisation.
          93 In the light of the other material, the reported crime statistics for the years 1995-1997, which suggest a significant increase over that period, should be treated by this Court with caution when it is suggested that changes in the incidence of the crime are such as to require a response in the sentencing practice of the courts. Nevertheless, the increase over the three year period is entitled to weight when deciding whether or not a guideline judgment is appropriate and, if so, what change with respect to past sentencing practice is appropriate.
          Victim Impact and Need for Guidelines
          94 The second body of material to which the Crown sought to make reference concerned the impact of armed robbery on victims. This included academic literature and the report of a clinical psychologist prepared for these proceedings. This material was said to establish the objective gravity and seriousness of the offence of armed robbery. There is no doubt that impact on victims is an aspect of the seriousness of an individual offence. General patterns of impact of the character referred to in the literature to which the Crown referred, including statistical surveys, confirm the seriousness of the offence.
          95 Plainly the actual impact in each particular case will vary and, appropriately, cause variations in the sentence imposed. This is not a manifestation of inconsistency. Rather, it represents the consistent application of a principle which varies in its import according to the circumstances.
          96 The surveys and other literature, to which the Crown referred, establish that armed robbery is perceived as a life threatening situation for a majority of victims. In one Australian survey 86 percent reported the robbery as being the most threatening experience they had ever had. Victims experienced both physiological and psychological problems. The physical effects included chronic nervousness, insomnia, nightmares, headaches, digestive problems. Psychological problems included generalised fear, depression, aggressiveness, mood changes. Sometimes the response was such as to develop into the clinical condition of post traumatic stress disorder.
          97 The surveys and academic literature confirm what trial judges and appellate judges would in any event know, namely, that armed robbery is a serious offence by reason, inter alia, of the significant impact it has on its victims. Indeed, the submissions on behalf of the Respondents to the Crown appeals did not, in contrast with criticism of the reliance by the Crown on crime statistics, challenge the materials put before the Court on the impact of the crime of armed robbery on its victims.
          98 The Crown also put before the Court a report prepared for purposes of this appeal by a clinical psychologist on the effects of emotional distress in the aftermath of personal trauma, not limited to cases of armed robbery. Nothing in the report was startling or controversial and no submissions were directed to it.
          99 This material confirms what this Court would in any event accept on the basis of commonsense and common knowledge. Armed robbery is not simply a crime against property. It is a crime against persons. Furthermore, the fear engendered by the perpetrator of this crime, together with the continued adverse effects on its victims, establish armed robbery to be a serious crime which requires condign punishment.
          Sentencing Statistics and Need for Guidelines
          100 The primary material before the court for the purpose of determining whether or not a guideline judgment is appropriate is to be found in the statistics kept by the Judicial Commission of New South Wales. Those statistics are available for all sentences imposed between December 1994 and October 1998 for offences against s97(1) of the Act, which extends to both armed robbery and robbery in company. In the period of almost four years covered by the statistics, there were a total of 835 cases. This is a significant number from which general conclusions as to sentencing practices can be drawn.
          101 The most important single aspect of the statistics is that of the 835 cases, a total of 688 (eighty-two percent) resulted in a full-time custodial sentence. That is to say, 147 cases, eighteen percent of the total, did not result in a full-time custodial sentence. Of those, the largest number, 64 (eight percent) resulted in a sentence of periodic detention and 55 (seven percent) resulted in a community service order.
          102 The Sentencing Information System of the Judicial Commission provides the following information for the 688 cases in which a sentence of imprisonment was imposed, divided into charts for the Full Terms and those for the Minimum or Fixed Terms.
          103

          At the other end of the spectrum, only seven individuals received more than half the maximum term of twenty years, i.e. more than ten years, as full terms. Furthermore, only about one-quarter of the entire sample of 688, being 163 cases, received full sentences of more than one-quarter of the maximum, i.e. more than five years.
          104 The chart with respect to the minimum or fixed terms is also revealing. The statistics do not allow the removal of the cases of fixed terms in order to look at the sentencing practice with respect to the relationship between minimum and full terms. On the basis of experience, it can be said that fixed terms are comparatively few in number. The statutory formula in the Sentencing Act 1989 (NSW) is for an additional term not exceeding one-third of the minimum term, unless “special circumstances” are found to exist.
          105 There are 307 cases out of the 688 total (45%) for which a minimum (or fixed) term of eighteen months or less was imposed. The application of the statutory ratio suggests that something of the order of that number of cases should have received full terms of twenty-four months or less (without making express allowance for the number of fixed terms). However, the full terms chart shows that only 73 individuals (not about 307) had sentences of twenty-four months or less imposed.
          106 For the next bracket up, the figure for minimum terms is 426 individuals who received two years or less for which the statutory ratio would imply a full term of thirty-two months. The Judicial Commission’s statistics provide full terms for thirty months and thirty-six months. In the former case, there were only 106 individuals sentenced and, in the latter case, 246.
          107 These statistics indicate that the “special circumstances”, which justify a ratio other than the statutory ratio, are found in the majority of cases. Furthermore, the figures strongly suggest that it is very often the case, as suggested in the particular instances before the court on this occasion, that trial judges have imposed a minimum term of significantly less than half of the full term.
          108 This is an offence for which the maximum penalty is twenty years. The statistics show that of the 688 persons sentenced to prison, sixty-two percent of the total receive an actual prison term (minimum or fixed) of two years or less. Eighty percent receive actual minimum or fixed terms of three years or less.
          109 These statistics indicate the following conclusions as to sentencing practices in the District Court for offences against s97(1) of the Crimes Act:
          (i) Non-custodial sentences are not confined to exceptional cases. The fact that 147 individuals received non-custodial sentences out of a total of 835 (i.e. 18%) is not consistent with an “exceptional case” test.
          (ii) The leniency of the sentencing pattern is also suggested in the full terms. (Over 60% of those who were in fact sentenced to prison received one fifth or less than the maximum, i.e. four years or less).
          (iii) The pattern of leniency is reinforced by the extent to which the statutory ratio between minimum and fixed terms has been altered so that over 60% of those who do receive prison sentences, receive a minimum or fixed term of two years or less.
          (iv) The large proportion of lenient sentences is also suggestive of inconsistency in sentencing practices between the group who receive full terms of 3-4 years or thereabouts and the group below that.
          110 These statistics strongly suggest both inconsistency in sentencing practice and systematic excessive leniency in the level of sentences. They justify the promulgation of a guideline judgment.
              CCA Cases
          111 One of the factors considered by the Court in Jurisic to be relevant to the determination of a need for a guideline judgment was the history of Crown appeals against sentence for the offence there under consideration, s52(A) of the Crimes Act. This Court had said, on a number of occasions, that that offence had to be treated as being more serious than it had been regarded. Notwithstanding those indications, there was a continued flow of, almost invariably successful, Crown appeals.
          112 The position with respect to s97(1) in this Court is not as unequivocal as the history with respect to the offence under consideration in Jurisic. Nevertheless, some aspects of this Court’s prior consideration of Crown appeals from sentences for s97(1) convictions are relevant.
          113 First, and most important, is the frequency with which the Court has stated that a non-custodial sentence for this offence could only be imposed in exceptional circumstances. As Hunt CJ at CL said in Roberts (1994) 73 A Crim R 306 at 308:
          “This Court has always made it clear that armed robbery is to be regarded as an offence of the utmost gravity which, except in the most exceptional circumstances, must carry a full time custodial sentence: Murray (unreported, 11 September 1986) at p5; Kingsbeer (unreported, 29 July 1988) at p7; Valentini (1989) 46 A Crim R 23 at 26; Readman (1990) 47 A Crim R 181 at 104-105; Diamond (unreported, 18 February 1993) at p2; Hetherington (unreported, 25 February 1993) at p4; Maddocks (unreported, 25 November 1993) at pp2, 6. That view has been said by this Court to be a sentencing principle resulting from its considered decisions to which sentencing judges should not merely pay lip service, Maddocks (at p6). Wood J, delivering the principal judgment, went on to say:
              ‘I would repeat, although I wonder why it is necessary that I should do so, that it is only in the most exceptional circumstances that anything other than a non-custodial sentence should be imposed for armed robbery. Necessarily, there will be cases which constitute an exception but they will be few and far between.’
                  With those observations, I express my complete agreement. I prefer the phrase “most exceptional circumstances” which he used to the phrase “wholly exceptional and unusual circumstances” subsequently employed in Crotty (unreported, NSWCCA 28 February 1994) at p5.”
          114 The cases referred to by Hunt CJ at CL indicate the frequency with which this Court had reiterated this basic proposition prior to 1994.
          115 The proposition, as summarised and reiterated in Roberts, has been applied subsequently in successful Crown appeals (Wright (1997) 93 ACrimR 48; Bragias NSWCCA 12 March 1997; Khoury NSWCCA 5 December 1994; Amohanga NSWCCA 25 May 1995).
          116 It has also been referred to in Crown appeals in which the Court has found the first instance sentence to be manifestly inadequate but, in the exercise of its discretion, decided to dismiss the appeal. (Sharpe NSWCCA 27 September 1994; Randall NSWCCA 19 September 1994; Kerr NSWCCA 26 August 1997; Gitt NSWCCA 18 May 1998).
          117 The Court has affirmed the basic proposition on occasions on which it has held that a finding of exceptional circumstances was open to the trial judge (Jones NSWCCA 15 April 1994; Georges NSWCCA 30 May 1996; Latouff NSWCCA 12 December 1996; Tocknell NSWCCA 28 May 1998).
          118 To use the words of Wood J from Maddocks, which Hunt CJ at CL quoted with approval in Roberts, non-custodial sentences should be “few and far between”. One thing that is clear from the Judicial Commission’s statistics is that non-custodial sentences imposed by trial courts for this offence are common. They cannot be described as “few and far between”. This long line of authority has not been reflected in the sentencing practice of trial judges.
          119 The distinction between the precedents in this Court on this offence, and the precedents considered in Jurisic, is that the Court has been prepared to uphold the finding of exceptional circumstances on a number of Crown appeals, whereas in Jurisic the disposition of appeals was almost all one way. However, prior to the present hearing, the pattern of sentencing by trial judges and this Court’s precedents on appeal, have never been reviewed in a systematic way. It is one of the advantages of a system of guideline judgments that this Court can review its own prior decisions from the perspective of ensuring consistency in the guidance it gives to trial judges. The pressures on the Court do not necessarily permit it to review its own decisions from this perspective in the normal course.
          120 In the course of its prior consideration, the Court has had occasion to make reference to the sentencing statistics. In Tocknell Hulme J, with whom Hidden J and Carruthers AJ agreed, said, with reference to the conclusion that a finding of exceptional circumstances was within the sentencing judge’s discretion:
          “In reaching that conclusion I place no weight on the sentencing statistics prepared by the Judicial Commission which were put before this Court. Insofar as some of those statistics suggest that something in the order of forty-seven percent of offenders guilty of the offences with which the respondent was charged and under twenty-one, had not been sentenced to full time custodial sentences, those statistics suggest to me that the judges of the District Court are not paying proper regard to the principles which this Court has laid down. If that impression be correct, and if the tendency continues, it may be that the traditional approach of this Court to appeals by the Crown may need to be reviewed.”
          121 The issue arose in a more acute form in the case of Ross (NSWCCA 14 May 1997). In that case, the sentence of a two year minimum term and a two year additional term was within the trial judge’s discretion to find “the case to be an exceptional one deserving leniency”. Studdert J and Levine J both agreed with the result, but with the following observations:
          “… The success of the appeal depended upon the Crown establishing that in all the circumstances the sentence was manifestly inadequate. In this regard the Crown case was not assisted by sentencing statistics produced to the court. Those statistics did not illustrate that the sentence here imposed was outside the range. Of course caution has to be used in resorting to such statistics and an appeal cannot be determined simply by reference to statistics of the nature it provided. …However, I am not persuaded that this Court should intervene.”
          122 Considerations of this character are plainly material. Particular scrutiny is appropriate where statistics indicate that the sentence under appeal is at the top or bottom of the range of sentences actually imposed, in circumstances where its objective characteristics did not indicate that it is appropriately so characterised. (See e.g. Bugmy (1991) 69 CLR 525 at 538; Bloomfield (1998) 44 NSWLR 734 at 739F). However, the court in Ross did not have before it a challenge to the usual level of sentencing in the way that that issue has been presented to the Court on this occasion. The judgments in Tocknell and Ross are precursors to the present judgment.
          123 The number of cases in which non-custodial sentences have been imposed indicates that the long established principle that such must be confined to “exceptional circumstances” has not been implemented by sentencing judges. The position is not quite as clear with respect to the level of the sentencing where custodial sentences are imposed. It is overwhelmingly likely that the approach to sentencing reflected in the imposition of non-custodial sentences has also been reflected in the length of custodial sentences when imposed. The sentencing statistics suggest that that is so.
          124 In Waldron (NSWCCA 3 March 1994) Hunt CJ at CL said (at p3):
          “This Court has on numerous occasions said that the range of sentences imposed for armed robberies, particularly serious ones, should be more deterrent than those which are in fact being imposed by the sentencing judges; see, for example Flack (12 December 1989); Smith (12 December 1984); Petrinovic (18 September 1990); Va (11 November 1993).” (3)
              See also Ellis (1993) 68 ACrimR 449 at 462; Lawson NSWCCA 12 December 1997.
          125 The sentencing statistics hitherto set out, indicate that the views expressed in Waldron have not been implemented. In part, this may have occurred because of the number of cases of findings of “exceptional circumstances” which this Court has endorsed on a basis other than the exercise of its discretion. A guideline judgment is required to ensure that the pattern of leniency and inconsistency indicated in the sentencing statistics is discontinued.
          Guideline Precedents
          126 The determination of an appropriate sentencing guideline for offences against s97(1) must commence with a recognition that the objective and subjective factors relevant to the exercise of the sentencing discretion, may vary over a wide range. See eg Brown (1989) 17 NSWLR 472 at 473-474. As I said in Jurisic with respect to s52A:
          “The nature of the offence is not such that the Court can devise a simple table in which indicative penalties are linked to a quantitative measure of the offence … What can be done, however, in case of an offence concerning a wide range of conduct which varies qualitatively rather than quantitatively, is to indicate in a general way the kind of case which would usually require a particular kind or level of sentence, whilst acknowledging that there will always be exceptional cases.” (35-36)
          127 A consideration of the practice of other criminal appellate courts, with respect to guidance for sentencing for armed robbery, suggests that such an approach is appropriate for this offence. I commence with the English cases.
          128 In 1972, the Court of Appeal (Criminal Division) heard together a number of severity appeals from convictions for armed robberies involving banks. The Court identified a “starting point” of fifteen years imprisonment for what the Court described as a “normal” bank robbery, where firearms were carried and no serious injury done. (Turner (1975) 61 CrAppR 67 at 91).
          129 In Gould (1983) 5 CrAppR(S) 72, the Court considered a number of cases of robbery of small business premises where the offenders were armed with guns, imitation guns, or knives. The Court confirmed the guideline in Turner as applicable to a serious type of case but added:
          “There are so many possible combinations of circumstance that it is difficult to give any precise indication of the so-called normal sentence for any particular type of robbery.” (75)
          130 The Court went on to identify a list of mitigating and aggravating circumstances. The Court found that the sentences in the three appeals before it (between five and eight years) were not excessive. It did not specify a starting point or sentencing range.
          131 The Court continued with this approach when it dealt with a number of Crown appeals in Attorney-General’s References Nos 3, 4, 8, 9, 10, 11 and 16 of 1990 (Dickson & Ors) (1991) 92 CrAppR 166. All these cases involved small shops. The Court allowed the appeals and substituted verdicts of between three and a half years and six years.
          132 The Court has subsequently referred to a seven year sentence in cases of this character as if it were a starting point. The case involved a very small bank branch and an imitation gun (Attorney-General’s Reference No 14 of 1991 (Reed) (1992) 13 CrAppR (S) 446). Lord Lane CJ said:
          “… it is very seldom that in these circumstances a sentence of much less than seven years would be appropriate, the purpose of the sentence being threefold: first of all to deter the offender from behaving in this way again; secondly to deter others from arming themselves with weapons and holding up banks; and thirdly, a matter which is sometimes overlooked, to punish the offender for having carried out this wicked crime and having put several people in fear of death.” (449)
          133 The seven year sentence in Reed was referred to as a starting point by Lord Taylor CJ in Attorney-General’s Reference No 7 of 1992 (Khan) (1993) 14 CrAppR (S) 122 at 125, 126, which also involved a small shop and an imitation pistol. His Lordship added:
          “It has to be realised that that type of shop is very often staffed by only one person who may be unable to defend himself or herself. It is unlikely that there will be any sophisticated security there, and it is a primary target for someone who wants to enrich himself quickly and successfully. It is therefore very important that the courts should indicate by the sentences passed that that type of offence will be punished severely.” (126)
          134 I should note that the Court of Appeal has subsequently issued a separate guideline judgment with respect to the use of firearms in a variety of offences (Avis (1998) 1 CrAppR 420).
          135 This reluctance in Gould to specify a range or starting point was based on the wide variation in the factual circumstances of individual offences. The reluctance to offer any guidance, where complete guidance is not feasible, has been criticised. (Ashworth “Techniques of Guidance on Sentencing” (1984) CrimLR 519 at 529-530; Ashworth Sentencing and Criminal Justice (2nd ed, 1995) at 30. It may be that that subsequent decisions are a response to this criticism.
          136 Guidance may be given by identifying certain commonly recurring categories of a specific offence. The New Zealand Court of Appeal has done this in Moananui (1983) NZLR 537. Guideline judgments in New Zealand are “bottom up” not “top down” guidelines, i.e. they purport to describe, rather than prescribe, sentencing practice. (See Hall “Reducing Disparity by Judicial Regulation: Sentencing Factors and Guideline Judgments” (1991) 14 NZ ULR 208 esp at 223-224).
          137 In Moananui the Court identified three categories:
              (i) Planned armed robbery at premises such as banks which endanger the safety of considerable numbers of people: which usually attract sentences of 6-8 years.
              (ii) armed robbery involving intrusion into dwelling houses, usually at night: examples given range from 4-6 years.
              (iii) armed robberies of smaller premises involving less extensive risk of injury and smaller sums of money: the range of sentences is lower. (Examples range from 2-5 years).
          138 A more overtly structured approach has been adopted in Hong Kong. In Mo Kwong Sang (1981) HKLR 610, the Court identified what it described as an “ordinary case” to be a starting point: armed robbery by an offender carrying a knife or other dangerous weapon, which was displayed to the victim. The Court said that an “appropriate” sentence for this case was five years. It indicated appropriate levels for two other common factual variations: invasion of private premises (six years) and physical violence (seven years).
          139 The Court also identified a list of aggravating and mitigating factors, noted that use of firearms would require more severe sentences, and affirmed the discretion of the sentencing judge to take into account subjective circumstances. The nature of the guidance provided by the appellate court was to:
          “… indicate the level of sentence for armed robbery with which we would not interfere. (611)
          140 I have noted above that Canadian criminal appellate courts have adopted a “starting point” approach to providing guidance to trial judges. This involves the identification of a typical case and the application of aggravating and mitigating factors. Courts in Alberta, Nova Scotia and New Brunswick have applied this approach to armed robbery. (See Young The Role of an Appellate Court in Developing Sentencing Guidelines Canadian Sentencing Commission (1988) p21ff).
          141 In Nova Scotia the appellate court has lain down a three year minimum sentence (Brennan and Jensen (1975) 11 NSR (2d) 84 at 88; Hingley (1977) NSR (2d) 541 at 544). The New Brunswick Court of Appeal has referred to three years as a normal sentence (Chaisson (1975) 24 CCC (2d) 159). In both cases the sentencing level was identified as appropriate to a young first offender, but there was little in the way of precision about other aspects of the offence or of the offender.
          142 In Johnas (1982) 2 CCC (3d) 490 the Alberta Court of Appeal focused on cases of robberies of “small commercial establishments, open at night for service to and convenience of the public, where often a single person is in charge” (494). The objective circumstances emphasised the degree of vulnerability. The category identified involved a threat of violence, but no actual violence. The Court identified the category as follows:
          “… what should be regarded as a fit sentence in Alberta for unsophisticated armed robbery of unprotected commercial outlets in the absence of actual physical harm to the victim and with modest or no success.” (495)
          143 The Court determined three years as a “starting point” and added:
          “… judicial reasoning as to a fit sentence for any offence must start with a norm for the type of offence involved. That norm is arrived at by comparisons to other cases, by experience, by the seriousness of the offence and by its prevalence. Having determined that norm, the court will look at the factors of mitigation and aggravation. Specific cases are not to be treated as precedents. Each is a unique mixture of aggravating and mitigating factors.” (499)
          144 In subsequent cases the Alberta Court of Appeal has specified five years as a starting point for bank robberies (Kurich (1982) 9 WCB 138) and four years for night deposit robberies (Hall (1983) 10 WCB 138).
          145 The Court of Criminal Appeal of Western Australia has, with respect to armed robbery, provided sentencing guidance in, what I have called above, a “bottom up” fashion: i.e. it is derived from the range of sentences actually imposed by trial judges. This involves identifying a tariff rather than fixing a starting point or sentencing range. It has been called “tariff sentencing”. (Thomas Principles of Sentencing (2nd ed, 1979) p29).
          146 In 1989, what the Court described as a “conventional armed robbery of a bank or similar premises”, attracted a sentence of “five to seven years or upwards” (Norman WACCA 1 February 1989). By 1997 these levels had been increased. The Court said:
          “… sentences have tended to firm up as a result of the increasing prevalence of the offence by giving greater weight to the requirements of deterrence and less weight to the antecedents and other matters personal to the offender … The offence of armed robbery has become significantly more prevalent since 1989 and sentences have firmed up in that period. At present, the range of sentences commonly imposed for a single offence, depending on the circumstances, would be from six to nine years.” (Miles (1997) 17 WAR 518 at 521 and 522).
          147 This level applies before a discount for a plea of guilty. The Court went on to note that where a blood filled syringe was used as a weapon, an additional one year’s imprisonment should normally be imposed, because of the additional impact on the victim (523-524).
          148 The Western Australia Court of Criminal Appeal has affirmed the range of 6-9 years in subsequent decisions (Moody 12 February 1998; Moulds 4 March 1998; Jeffree 1 May 1998). In Jeffree, the Court further clarified the application of the range which was described in Miles as “commonly imposed”:
          “Offences of this type commonly do not progress beyond a threat or violence of a minor kind. Those which do could be characterised as amongst the worst type of case and would be outside the usual range.”
          149 The range of sentences for this offence in Western Australia - six to nine years - must be understood in the light of a discount for a guilty plea (said to be 20-35%, Miles at 521) and the statutory framework for sentencing, which differs from that in New South Wales. By s93 of the Sentencing Act 1995 (WA), the range of six to nine years could permit release on parole after, in broad terms, two to four years. This, again in broad terms, would equate to a minimum term under the Sentencing Act 1989 (NSW).
          150 As I have indicated above, the Full Court of the Supreme Court of South Australia has promulgated “sentencing standards” in certain cases. In Cadd (1997) 94 ACrimR 466 at 480 Doyle CJ identified the authorities which had set such standards for the offence of armed robbery. He referred to Spiero (1979) 22 SASR 543 as setting a standard and to Dube (1987) 46 SASR 118 as a case in which the Court contemplated, but refrained from, increasing the level.
          151 In Dube the Court considered a Crown submission that this standard should be increased because of the increased prevalence of the offence. The submissions made are strikingly similar to those before this Court in the seven cases now before it. King CJ noted:
          “The prosecution adduced evidence before the learned sentencing judge as to the adverse affect of armed robbery upon the victims and innocent people who become embroiled in it. There was also evidence as to the financial cost of armed robberies to financial institutions and to the community generally … The evidence led by the prosecution in this case merely reinforces the view long held on this Court that armed robbery presents a serious threat to the wellbeing of the community and that punishments imposed must reflect that fact …
              The prosecution produced evidence before the learned sentencing judge of the increasing prevalence of this crime during the last seven years. It is unnecessary to recount the details of this evidence; it is sufficient to say that it establishes that the increase has been marked and is apparently continuing. The much discussed question of the effectiveness of imprisonment as a deterrent to crime, and in particular of the effectiveness of increased levels of punishment, was adverted to during argument. I think that it must be conceded that there is no proven correlation between the level of punishment and the incidence of crime and that there is no clear evidence that increased levels of punishment have any effect upon the prevalence of crime. Nevertheless the criminal justice system has always proceeded upon the assumption that punishment deters and that the proper response to increased prevalence of crime of a particular type is to increase the level of punishment for that crime. I think that courts have to make the assumption that the punishments which they impose operate as a deterrent. That being so, I think that it follows that the proper response, and the response which is expected by the community at large, to the increased prevalence of serious crime is increased punishment for that crime.” (119-120)
          152 The reason the Court refrained from formally increasing the standard was because of the then recently amended legislative provision on the treatment of remissions, which would have the effect of increasing sentences.
          153 Spiero involved an armed robbery of a chemist shop by three offenders. Firearms were used. The sentence of eight years fixed in that case was treated subsequently as a standard. The appropriate range has been said to be eight to twelve years, in all but “exceptional cases”. (See Fermaner (1994) 72 ACrimR 138 at 139; 61 SASR 447 at 448 based on Prendergast (1988) 147 LSJS 486 at 486-488).
          154 Fermaner was heard together with three other Crown appeals from sentences for the offence of armed robbery: Bini, Meo and Ramsden, each SACCA 21 March 1994 (unreported). The four cases are analysed in Hinton “Principled Sentencing in South Australia - Armed Robbery and the Quest for Consistency” (1995) Flinders J L.Reform 18.
          155 Three of the four cases have similar objective circumstances: robbery of a service station; offenders aged from 24 to 36 years; armed with a knife, a screwdriver and a small axe; a few hundred dollars taken; an early plea of guilty. There were differences in terms of prior record and effect on victim. The Court of Criminal Appeal set aside the sentences below and imposed sentences of five years four months (on top of five months served before sentence); with a non-parole period of four years; six years with a non-parole period of four years and 5 years 8 months, with a non-parole period of three years.
          156 In these cases (Fermaner, Bini and Ramdsen) eight years was treated as a starting point. In Fermaner the Court indicated that this would have been the appropriate head sentence but for the early plea of guilty and the period of five months in custody before sentence (140). This was reiterated in Bini and Ramsden. The plea of guilty led to a discount of two years (Hinton supra p27).
          157 Since the reaffirmation of the eight year starting point in these four 1994 cases, a system of truth in sentencing was introduced by the Statutes Amendment (Truth in Sentencing) Act 1994 (SA). As a result, the accrual of remissions, which the Court had referred to in Dube, would no longer be taken into account in sentencing. This would require a reduction of the appropriate range hitherto identified for armed robbery by the extent to which sentences had been inflated to take account of remissions. A reduction of one third would accord with the transitional provisions of the Statutes Amendment (Truth in Sentencing) Act (See Hinton supra pp25-26).
          158 This suggested a reduction in the eight to twelve years for the head sentence, to about six to eight years. This has since been affirmed by the South Australian Court of Criminal Appeal. (Drumgoon SACCA 20 November 1995; Branscherd SACCA 22 May 1996). The starting point of six years in South Australia was before a plea of guilty and other discounts (eg co-operation with the police). (See Drumgoon supra).
          159 The Queensland Court of Appeal reviewed a number of prior decisions in Hammond (1996) 92 ACrimR 450 at 456ff. The case involved pleas of guilty to three robberies of service stations by an offender armed with a steering wheel lock. After referring to the need for consistency with the prior decisions it reviewed, the Court imposed a sentence of five years with a recommendation for release on parole after two years.
          160 In a subsequent case involving a single armed robbery of a service station when armed with a knife, by an offender who pleaded guilty, the review in Hammond was referred to, the severity appeal allowed and a sentence of four years substituted. (Morton (1997) 95 ACrimR 381).
          A Guideline for New South Wales
          161 In the cases now before this Court, the Crown has propounded a starting point for New South Wales of six years full term, in circumstances where there is a plea of guilty.
          162 It appears from the cases that come to this Court, including the present proceedings, that there is a category of case which is sufficiently common for purposes of determining a guideline:
              (i) Young offender with no or little criminal history
              (ii) Weapon like a knife, capable of killing or inflicting serious injury
              (iii) Limited degree of planning
              (iv) Limited, if any, actual violence but a real threat thereof
              (v) Victim in a vulnerable position such as a shopkeeper or taxi driver
              (vi) Small amount taken
              (vii) Plea of guilty, the significance of which is limited by a strong Crown case.
          163 Whilst it is possible to determine a starting point in a case of this kind, i.e. a sentence of X years imprisonment, I do not believe that the Court should do so. Rather, I propose the Court should identify a narrow sentencing range within which this Court would expect sentences in such cases to fall.
          164 There are two principal reasons why a sentencing range is appropriate for this offence:
              (i) The seven characteristics identified above do not represent the full range of factors relevant to the sentencing exercise.
              (ii) Many of the seven identified characteristics contain within themselves an inherent variability, eg different kinds of knives or weapons in (ii); extent of “limited actual violence” in (iv); degree of vulnerability in (v); amount in (vi).
          165 In my opinion sentences for an offence of the character identified above should generally fall between four and five years for the full term. I have arrived at this figure after drawing on the collective knowledge of the other four members of the Court with respect to sentence ranges. I have also reviewed the sentences which this Court has imposed on occasions when it has intervened, including in Crown appeals where the principle of double jeopardy applies. The proposed range is broadly consistent with this body of prior decisions in this Court.
          166 Cases consistent with the range are Sandaford 10 September 1995; Sedgwick 7 July 1992; Martin 19 February 1993; Sarkas 8 September 1992; Walters and Watkins 2 June 1994; Achurch & Brady 8 July 1994; Willett 6 November 1995; Maretta 30 October 1996; Bragias 12 March 1997; Basaga & Weleilakeba 9 August 1998.
          167 Cases above the range to varying degrees are Salameh 12 March 1991; Sneddon 28 March 1991; Marsden 15 April 1993; Antoce 22 April 1993; Walker 21 February 1994; Pettit 9 March 1994; Sneddon 28 March 1994; Nguyen 14 April 1994, Burger 19 July 1994; Rafter 23 September 1994; Harborne 12 October 1994; Murray 7 November 1995; ; Putescu 20 December 1996; White 7 April 1997; Lowe 24 April 1997.
          168 Cases below the range being sentences between 2 and 3½ years are Bateman 15 July 1993; Roberts 24 May 1994; Amohanga 25 May 1995; Davies 29 November 1995; Wright 28 February 1997; Fisher 8 September 1998; Wightman 2 November 1998.
          169 Aggravating and mitigating factors will justify a sentence below or above the range, as this Court’s prior decisions indicate. The narrow range is a starting point.
          170 In addition to factors which may arise in any case eg youth, offender’s criminal record, cooperation with authorities, guilty plea in the absence of a strong case, rehabilitation efforts, offence committed whilst on bail etc, a number of circumstances are particular to the offence of armed robbery. These include:
              (i) Nature of the weapon
              (ii) Vulnerability of the victim
              (iii) Position on a scale of impulsiveness/planning
              (iv) Intensity of threat, or actual use, of force
              (v) Number of offenders
              (vi) Amount taken
              (vii) Effect on victim(s)
          171 The Respondents to the Crown Appeals also submitted that this Court should now declare that the fact that the offender was addicted to drugs, and committed the crime to obtain money to feed his or her habit, is a mitigating circumstance. This would require the Court to overturn a long line of prior decisions.
          172 In 1990, the Australian Institute of Criminology published a detailed analysis of the practice of this Court with respect to sentencing for robbery over the course of a ten year period. That report made it clear that the Court had consistently regarded drug addiction as a mere explanation, not an excuse. (Potas Sentencing Robbers in New South Wales (1990) at pp142, 163).
          173 As a joint judgment of the Court put it in Valentini (1989) 46 ACrimR 23 at 25:
          “This Court has said on countless occasions that addiction to heroin is not to be considered as a factor for the reduction of what would otherwise be an appropriate sentence for the nature of the offences which have been committed. It serves, however, to provide an explanation for the commission of the offences.”
              This has been reiterated in a number of later cases (eg Stanford NSWCCA 23 November 1993; Atkins NSWCCA 27 May 1998).
          174 Drug addiction is one of the circumstances of a particular offence that is relevant to the sentencing exercise. It may, for example, be pertinent to the issue of impulsiveness/planning or to the weight to be given to rehabilitation in a particular case.
          175 In Engert (1995) 84 ACrimR 67, Gleeson CJ emphasised the complexity of the sentencing task and the multiplicity of the, sometimes conflicting, objectives it must serve. His Honour concluded:
          “It is therefore erroneous in principle to approach the law of sentencing as though automatic consequences follow from the presence or absence of particular factual circumstances. In every case, what is called for is the making of a discretionary decision in the light of the circumstances of the individual case, and in the light of the purposes to be served by the sentencing exercise.” (68)
          176 His Honour went on to say that the existence of a causal relationship between a mental disorder and the commission of an offence does not automatically produce the result that the offender will receive a lesser sentence (at 71). Nor, in my opinion, should the existence of a causal relationship between drug addiction and the commission of an offence.
          177 The circumstances in which motive may be a mitigating factor should, in my opinion, be confined to cases in which motive impinges upon the moral culpability of the offender. This can include mental, emotional or medical problems or impulsive conduct. In any event the degree to which motive may be pertinent must depend on the whole of the circumstances. (White NSWCA 23 June 1998 at 24).
          178 As noted above, it has long been the position taken in this State that addiction is not, of itself, a mitigating circumstance. This is also the position in England, (Brewster (1998) 1 CrAppR 220 at 226) Canada (Johnas (1982) 2 CCC (3d) 490 at 497). It was submitted for the Respondents to the Crown appeals that authorities in other Australian states indicate a different approach.
          179 Reliance was placed on Nolan (1998) VICCA 135 (2 December 1998). This was a majority decision to allow a severity appal. One member of the majority, Buchanan JA, referred to drug addiction in terms of “mitigating circumstances” (par 15). The other member of the majority, Butt JA, did not adopt that terminology, but rather that of “explanation” (par 19).
          180 Buchanan JA referred without disapproval to Halewyn (1984) 12 ACrimR 202 at 203 where Young CJ said:
          “… it has been said a number of times in this Court that the fact that an armed robbery is committed to support a drug habit is of little consequence in mitigating the sentence to be imposed.”
          181 Buchanan JA also referred to Bouchard (1996) 84 ACrimR 499 where at 501-502, Callaway JA quoted Hallwyn but identified a case of “a man crazed by a drug craving who committed an offence on the spur of the moment and later repented”.
          182 Nolan was such a case. It should be understood as an authority on the issue of planning/impulsiveness, and that the impulsiveness was the “mitigating circumstance”, rather than the drug addiction itself.
          183 I do not understand there to be any difference between the position in New South Wales and that in Victoria in this regard.
          184 In Douglas (1995) 56 FCR 465 the Full Court of the Federal Court heard an appeal from the Supreme Court for the Australian Capital Territory. In a joint judgment, Von Doussa, Higgins and R D Nicholson JJ said (at 470):
          “It is, of course, not a mitigating factor that a person commits a crime to feed a drug addiction. However, that is not a principle of universal application. As Jenkinson J noted, at 105, in Talbot (1992) 34 FCR 100, the moral and legal fault for the acquisition of the addiction has to be considered:
              ‘… evaluation of moral culpability remains in my opinion as fundamental to one system of punishment for crime as evaluation of the degree of harm caused or threatened by the criminal act or actor.’
                  The age of the offender when he or she becomes addicted and the degree of judgment open to them at that age is thus relevant in evaluating the extent to which they should be punished for consequential criminal conduct.”
          185 Douglas involved an addiction that commenced at age eight and was total by age 13. This is an extreme case of little assistance in establishing a general principle that addiction lessens moral culpability. On the contrary, it affirms that self induced addiction at an age of rational choice establishes moral culpability for the predictable consequences of that choice. That was the very distinction made by the Full Court in Talbot (1992) 34 FCR 100 at 105-106.
          186 In that case, the Court took into account an addiction that was caused by medical treatment “to which his consent was at best merely formal” (105). (See also the Victorian decision in Redenbach (1990) 52 ACrimR 95 at 99). The Court referred with approval to, whilst distinguishing, the reasons of King CJ in Spiero (1979) 22 SASR 543:
          “One feels sympathy for the person who has become entangled in drug addiction, but the courts cannot treat addiction as an excuse, or even a mitigating factor, in relation to serious crime. Those who are addicted to drugs must understand that if they allow their addiction to lead them into serious crime, they must expect to receive the same serious punishment as would be received by others.” (549)
          187 In Terizakis (1986) 41 SASR 252 at 256, the Full Court reiterated this reasoning. O’Loughlin J added:
          “Despite the cravings caused by the addiction, the majority of drug addicted persons still retain, in varying measures, the ability to distinguish between right and wrong and the ability to comprehend that the consequence of armed robbery will be a lengthy term of imprisonment. To give way, out of sympathy and concern, by imposing lesser terms of imprisonment that would otherwise be imposed, might appear to be an act of humanity, but it will do nothing to stamp out the drug trade and worse, it might even give some indirect encouragement to would-be offenders.” (256)
          188 Another authority on which the Respondents to the Crown appeals rely is the decision of the Queensland Court of Appeal in Hammond (1996) 92 ACrimR 450.
          189 In Rosenberger (1996) 76 ACrimR, that Court had indicated that intoxication, whether by alcohol or drugs, would not mitigate penalty save in the case where the original addiction did not involve a free choice, (as in Talbot or Redenbach supra, the latter being quoted in Rosenberger).
          190 In Hammond the Court distinguished Rosenberger and said:
          “The true relevance of drug addiction as a factor contributing to the commission of crime and its effect in the sentencing process has never been adequately explained. In our view it is a factor that may help an offender to the extent of showing that his or her descent into the crime in question was a secondary consequence of desperation produced by a human weakness rather than a primary choice. It may be inferred in many cases (and the present case is very much in point) that the offender would almost certainly never have become a thief but for loss of control of an ordered life through drug addiction. The proposition that the results of drug addiction are self-inflicted is half true and therefore dangerous. The offender must of course be held responsible for his or her own actions. The drug addiction is not an excuse; but it is a factor that may tell the court that the real weakness of character is that of a drug addict rather than that of a robber. That may be by no means inconsequential. It is however a two-edged factor; it may also tell the court that rehabilitation is going to be difficult.
              Just how these factors are to be applied to particular cases had not been spelled out, and it may be undesirable to attempt to do so. In our view the fact that drug addiction has led to criminal activity can properly lead to a wider understanding of the offender’s character and motivation for the crime. Depending on the circumstances of the particular case this may sometimes show the offender to be less deserving of condemnation than the primary facts, without more, might suggest.” (455-456)
          191 In its conclusion the Court returned to this theme:
          “Addiction to drugs at the time of the offence is not an excuse but may tell the court that the primary weakness of character is that of an addict rather than that of a robber. Such a factor tends to be mentioned along with an unfortunate family life as an explanation of the offender’s descent into bad habits. When the original problem was not something for which the offender was responsible, it is natural to feel some sympathy for the offender, though this must not obscure the offender’s direct responsibility for the ultimate criminal act he or she chose to do.” (467)
          192 This latter passage appears in the context of the Court listing aggravating and mitigating factors. It is noticeable that addiction is not identified as a mitigating factor. In neither passage is addiction, of itself, treated as a basis for imposing a lower sentence than would otherwise be imposed. In both passages it is emphasised that addiction is “not an excuse”. In both passages the offender’s responsibility for his or her conduct is mentioned.
          193 In my opinion, Hammond affirms that drug addiction is a relevant circumstance but is not, of itself, a mitigating factor. If, contrary to this conclusion, the remarks in Hamilton were intended to suggest otherwise - and some of the language chosen may be so interpreted - then they should be regarded as anomalous and should not be followed.
          194 The authorities are against the proposition that drug addiction should, of itself, be accepted as a mitigating factor. There is authority that where the original addiction was not a willed act, that may be taken into account by way of mitigation. The authority does not go beyond that.
          195 It is in the context of this line of authority that the submissions made for Respondents to the Crown appeals to the effect drug addiction, at least where it can be shown to be causally related to the commission of an offence, should now be accepted to be a mitigating circumstance.
          196 It was submitted that the degree of moral culpability of a particular offender is diminished by addiction. Evidence was put before the Court that there is, at least in some cases, a genetic predisposition to addiction and that addiction generally is not simply a state of mind but has a neurobiological or physiological base. It was put that an addict’s decision to perform a criminal act was not “a completely free choice”.
          197 In my opinion drug addicts who commit crime should not be added to the list of victims. Their degree of moral culpability will vary, just as it varies for individuals who are not affected by addiction. There are a number of aspects of the relationship between drug addiction and crime which indicate that moral choices are made.
          198 First is the original decision to experiment with drugs which, in the usual case, is a completely free choice. The addictive quality of drugs, together with the anti-social behaviour which so commonly results from addiction, is so widely known that persons who choose a course of addiction must be treated as choosing its consequences.
          199 Secondly, the submissions in this Court were in error in identifying the relevant conduct as the craving associated with withdrawal. The material presented to the Court did not suggest that the choice faced by addicts was between this negative feeling and the need for money to allay it. Rather the choice may often be the desire for the positive feeling said to be associated with a drug induced euphoria. The desire to bring about that state of “well-being” is, relevantly, a moral choice.
          200 Thirdly, nothing in either the process of addiction or its neuro-biological and physiological basis, leads ineluctably to the commission of crime, let alone the commission of crimes of violence against persons, such as armed robbery. Not all persons who suffer from addiction behave in this way. Those that do so, make a choice.
          201 Finally, individuals do emerge from addiction. They do so with difficulty and generally need significant amounts of help. The decision to persist with an addiction, rather than to seek assistance, is also a choice.
          202 There is no warrant, in my opinion, to assess a crime induced by a need for funds to feed a drug addiction, as being lower in the scale of moral culpability than other perceived requirements for money.
          203 Counsel making these submissions was driven to accept the proposition that an addiction to gambling, with its attendant need for money, could similarly be the basis for a claim for mitigation. He went further and indicated that an elevated sex drive which also had a physiological basis, could be a mitigating factor for the commission of a rape or for the conduct of a paedophile. In my opinion all these submissions should be rejected.
          204 It was also submitted that the principle of general deterrence should be given less weight in the context of offenders who commit their crimes for the purposes of assuaging a drug addiction. In this submission, reliance was placed on the often expressed doubts about the direct effect on potential offenders of increases in penalties imposed. Particular reliance was placed on the driven nature of a drug addicts, many of whom engage in high risk activity, like exchange of needles notwithstanding the risk of HIV infection. This kind of submission has been made many times before and I have quoted a few instances above. It has always been rejected. This Court should do so again.
          205 General deterrence always operates at the margin. Some people will continue to engage in criminal conduct notwithstanding the level of, or increases in the level of, the penalties they suffer. However, some people will be deterred. It is not to the point that some addicts engage in high risk activities. It would be necessary to establish that all addicts do so. Neither the submissions, nor the materials in support, suggest anything of this character.
          206 I attach particular significance to the impact that acknowledgment of drug addiction as a mitigating factor would have on drug use in the community. The sentencing practices of the courts are part of the anti-drug message, which the community as a whole has indicated that it wishes to give to actual and potential users of illegal drugs. Accepting drug addiction as a mitigating factor for the commission of crimes of violence would significantly attenuate that message. The concept that committing crimes in order to obtain monies to buy an illegal substance is in some way less deserving of punishment than the commission of the same crime for the obtaining of monies for some other, but legal, purpose is perverse.
          207 It may very well be the fact that increased possibility of detection has greater effect by way of deterrence than increased punishment. There is no warrant, however, for the Courts abandoning reliance on the latter. In any event the two propositions are related. It is only because detection, when it occurs, leads to a level of punishment, that increases in detection have their deterrent effect.
          208 It may very well be that the criminal justice system has a modest role to play in the control of drug addiction. But however modest that role may be, it must be performed in accordance with the basic structure of the criminal sentencing process. At the level of a structure deeply embedded in our society, not merely at the level of an individual’s calculus of risks and benefits for specific conduct, the criminal justice system is now, and has always been, based on the proposition that punishment deters and, within limits of tolerance, increased punishment has a corresponding effect by way of deterrence. This Court should not change such a longstanding assumption. Legislation would be required to alter the common law in this way.
          209 I reiterate that the process of imposing penalties for the commission of crimes, has its primary deterrent effect through its operation as a structural phenomenon of the criminal justice system. That is not capable of being assessed from the perspective of what particular penalties, or increases in penalty, may have in the case of individuals.
          210 In any event, the reasons for the guideline propounded in this judgment do not relate merely to an increase in the size of penalty. The guideline is particularly directed to overcoming the very significant proportion of cases in which non custodial sentences have been imposed. Henceforth, such sentences should be restricted to the exceptional cases to which the authorities have always referred.
          211 The deterrent effect of a sharp reduction in the proportion of non-custodial sentences which, as a result of this judgment, become custodial sentences, may well be much more significant than the deterrent effect from an increase in the level of custodial sentences. That is not to say that the latter does not have some deterrent effect but the sentencing practices, which I have analysed above, have been such as to significantly attenuate the deterrent effect of sentencing for the offence of armed robbery in the past. It is my opinion that that should change.
          212 I have now read the additional observations of Wood CJ at CL. I agree with his Honour’s judgment.
          **********


          IN THE COURT OF
          CRIMINAL APPEAL

          No. 60559/98
          60558/98
          60561/98
          60746/98
          60596/98
          60595/98
          60511/98
        SPIGELMAN CJ
        WOOD CJ at CL
        NEWMAN J
        HULME J
        SIMPSON J
          Wednesday 12 May 1999
            REGINA v Paul Anthony HENRY
            REGINA v Stephen Anthony BARBER
            REGINA v Hoai Vinh TRAN
            REGINA v Troy David SILVER
            REGINA v Theo TSOUKATOS
            REGINA v Bill KYROGLOU
            REGINA v John David JENKINS
            GUIDELINE JUDGMENT


            213 WOOD CJ at CL: I have read in draft the reasons of Spigelman CJ. I agree with the reasons for delivering a guideline judgment in relation to the offences before the Court, and with the guidelines proposed.
            214 I wish, however, to examine in a little more detail the submission that this Court should now declare that the fact that an offender was addicted to drugs, and committed an armed robbery in order to obtain money to support a drug habit, is a mitigating circumstance. As Spigelman CJ has observed, this would require the Court to overturn a long line of prior decisions, and to part from the position adopted in other jurisdictions, both within this country and elsewhere.
            Drug Addiction
            215 It cannot be gainsaid that very many offences of armed robbery are committed because of an addiction to drugs. All of the respondents to the Crown appeals presently before the Court were motivated by a drug habit. The New South Wales Bureau of Crime Statistics and Research reported in 1987 that:
            “drug (usually heroin) dependency appears to be an important ingredient in robbery offending, and in particular in the pattern of offending of those with multiple robbery convictions.”
            (New South Wales Bureau of Crime Statistics and Research, Robbery: Final Report (1987) at 88. When the Bureau interviewed convicted armed robbers, it found that drug use was a motivating factor in 70% of robberies (ibid at 93.)
            216 The acceptance of the link between drug dependency and many forms of criminal activity has recently led the State of New South Wales to create a Drug Court, with a special regime for dealing with certain categories of offenders, although not those charged with offences “involving violent conduct”: Drug Court Act 1998 (NSW) S 5(2)(b).
            217 The question whether drug addiction should operate as a mitigating factor when sentencing an offender for an offence under S 97(1) Crimes Act, or for similar offences, has been considered by this Court upon numerous occasions. The attitude taken to date is accurately reflected in Valentini (1989) 46 A Crim R 23; and in Ellis (1993) 68 A Crim R 449.
            218 In Valentini, the Court (comprised by Maxwell, Carruthers and Loveday JJ) said, (at 25):
            “This Court has said on countless occasions that addiction to heroin is not to be considered as a factor for the reduction of what would otherwise be an appropriate sentence for the nature of the offences which have been committed. It serves, however, to provide an explanation for the commission of the offences.”
            219 In Ellis, Hunt CJ at CL said (at 462):
            “The maximum sentence enacted by the legislature indicates that it expects the courts to impose severe sentences for these offences. I believe that the community has the same expectation. As was said by this Court in a bank robbery case (Petrinovic (unreported, 18 September 1990) at 9) the citizens of this community are sick and tired of armed robberies conducted by criminals in order to feed their drug habits, and this Court will support judges who recognise these as serious crimes and who impose heavy sentences for them.”
            220 Although Kirby P (at 459) expressed his concern in Ellis at the possibility of the Court being diverted from its legal duty, by “appeals to populist notions of what the Court conceives the community to be ‘sick and tired of’”, and dissented in the decision of the majority in dismissing the appeal, that was attributable to the comparison that he thought appropriate with the sentences that had been imposed upon separate offenders in analogously similar cases. His Honour did advert to the fact that:
            “people addicted to illegal drugs frequently have had resort to armed robberies in an attempt to secure the funds to feed the habit made expensive by its illegality”
                but his judgment provides no encouragement for the view that such an addiction should be regarded as a factor in mitigation.
            221 In Begnell (Court of Criminal Appeal New South Wales 26 November 1992) his Honour observed (at 4):
            “The courts cannot, in the face of the provisions of the statutes which criminalise the conduct of the use of prohibited drugs, allow that use to become an excuse for the kind of conduct which results in the charges and convictions that bring the applicant before us today. On the other hand, it is necessary, as it seems to me, to take into account such a consideration in looking at the criminality of the action in its totality and in considering its seriousness.”
            222 I do not understand his Honour, by the qualification contained in the second sentence of this passage, to be suggesting that the fact that an offence was committed to cater for a drug habit, should be regarded as a factor of mitigation. The contrary was, in fact, suggested by his Honour’s comments in Bradley (Court of Criminal Appeal New South Wales 26 October 1993) in a case of assault and rob involving a bag snatch from an elderly lady in a suburban street by an offender with a cocaine dependence, (at 5):
            “… her past addiction to cocaine was not an excuse for the action of the kind which occurred on this occasion. There are organisations and individuals in the community who will help people with this form of addiction. It would be completely intolerable if people with an addiction of this kind could resort to acts of violence against elderly members of the community or indeed anyone else. The courts must make it plain that such conduct will not be tolerated.”
            223 Similarly, in Vidler (Court of Criminal Appeal New South Wales 10 April 1986) Street CJ, with whom Lee and Lusher JJ agreed, said that:
            “Drug addiction with the consequently generated cash need … may explain criminal activities but it cannot excuse them so as to result in diminution in sentences proper to be passed for the offences under consideration”.
            224 In Crotty (Court of Criminal Appeal New South Wales 28 February 1994) where the offence involved was one of armed robbery of a taxi-driver, Sully J said (at p6):
            “Drug induced crime, particularly drug induced armed robbery, will not be regarded by the Courts as being necessarily mitigated by the particular drug and alcohol problem. It needs to be said yet again that ordinary decent members of the community who look, and rightly look, to the Courts for protection of their persons and property against assault and robbery are fed up with the notion that somehow or other a person who assaults them, who steals their property, who breaks into and ransacks their homes, is somehow not to be blamed for what he or she has done because of the presence of some drug or alcohol or like sociological problem invoked to explain what has been done in a particular case. I do not suggest, of course, that the presence of a drug or alcohol problem is not a relevant consideration.
                All I am seeking to say is that, as with the bare fact of youth, so also with the bare fact of drug and alcohol inducement, it must not be allowed to become a cloak of convenience behind which people who are in every relevant sense adults can shelter from the responsibility for their serious breaches of criminal law.”
            225 More recently it was said, by McInerney J, with whom Bruce J agreed, in Jebara (Court of Criminal Appeal New South Wales 2 August 1995):
            “It has been pointed out that whatever sympathy the courts may feel for a person who is subject to a compulsive and expensive habit, it cannot be made an excuse for crime. The community has to be protected against criminal action, whatever the motivation may be. In R v Martin (Court of Criminal Appeal New South Wales 19 March 1992 unreported) Hunt CJ at CL referring to such a habit providing a reason for the commission of offences, stated it provided no excuse in mitigation. At best, such a circumstance avoids a finding of aggravation by reason of the offence being committed for pure greed."
            226 This approach, which has been referred to as a “hard line” approach (Potas, Sentencing Robbers in New South Wales 1990 at 242, 163 can also be seen in decisions such as: Atkins (Court of Criminal Appeal New South Wales 27 May 1999); Hines (Court of Criminal Appeal New South Wales 25 May1998); Salameh (Court of Criminal Appeal New South Wales 9 June 1994); Crotty (Court of Criminal Appeal New South Wales 28 February 1994 unreported); Stanford (Court of Criminal Appeal New South Wales 23 November 1993); Jansz (Court of Criminal Appeal New South Wales 30 September 1993); Shinner (Court of Criminal Appeal New South Wales 10 July, 1992); and Cottier (Court of Criminal Appeal New South Wales 18 December 1990).
            227 A departure from this line of authority can be seen in the observation of Hulme J, in Murray (Court of Criminal Appeal New South Wales 7 November 1995) where his Honour said that:
            “the fact that (the applicant’s second criminal) career was inspired by a need for funds to support his heroin habit, is, to my mind, a mitigating factor”.
            228 He was, however, in the minority on this point, Grove J, with whom Hunt CJ at CL agreed, saying:
            “I would not regard an explanation founded in a need to support a heroin habit as a mitigating factor in regard to these offences. No doubt the habit offers explanation but in my view, it does no more than that.”
            229 A similar stand has been taken elsewhere: See, for example, Lawrence (1988) 10 Cr App 464; Brewster (1998) 1 Cr APP R 220, at 226, and Johnas (1982) 2 CCC (3d) 490, at 497; and Terizakis (1986) 41 SASR 252 at 256 and also at 252 where Zelling ACJ said:
            “It cannot be too strongly stated that heroin addiction and the need to obtain money in furtherance of it, is not a matter in mitigation in charges of robbery under arms.”
            230 Reference may also be made to Spiero (1979) 22 SASR 543 at 5488-549, where King CJ with whose reasons for judgment Walters and White JJ concurred, said:
            “one feels sympathy for a person who has become entangled in drug addiction but the Court cannot treat an addiction as an excuse, or even a mitigating factor, in relation to serious crime. Those who are addicted to drugs must understand that if they allow their addiction to lead them into serious crime, they must expect to receive the same severe punishment as would be received by others.”
            231 In Halewyn (1984) 12 A Crim R 202, Young CJ, in whose judgment Kaye and Beach JJ concurred, said (at 203):
            “The offence thus committed was a very serious one which, as I have said, is all too prevalent in this community. It is an offence which the courts are bound to deal with by imposing substantial sentences, and it has been said a number of times in this Court that the fact that an armed robbery is committed to support a drug habit is of little consequence in mitigating the sentence to be imposed.”
            232 In Talbot (1992) 34 FCR 100, Jenkinson J, with whose reasons O’Loughlin and Higgins JJ agreed, said at 105-106:
            “What in my opinion justifies the denial to the addict of leniency in recognition of the relative strength of his temptation to obtain his drug by armed robbery are on the one hand the magnitude of the community’s need of protection from that offence and on the other hand the moral and legal fault which originated the addiction. It seems unfashionable - and, having regard to recently acquired knowledge of molecular biology, perhaps imprudent - to express moral judgments in sentencing criminals. But evaluation of moral culpability remains in my opinion as fundamental to our system of punishment for crime as evaluation of the degree of harm caused or threatened by the criminal act or actor. So much I think the reasons of Mason CJ, Brennan, Dawson and Toohey JJ for their judgment in Veen v The Queen (No 2) (1988) CLR 465 make clear, particularly in their quotation (at 473-474) from an article by C S Lewis (Even more worthy of judicial citation is Lewis’ The Abolition of Man (2nd ed 1946).”
            233 Their Honours did, however, in this case note the special circumstance where the addiction of the offender originated in the administration of a potentially addictive substance in the course of medical treatment, to which his or her consent was, at best, merely formal. The moral culpability of such an offender was said to be different from that of the offender whose addiction originated in the voluntary administering of a substance that he knew (or I would add, should have known) to be addictive, and the use of which is proscribed by the criminal law.
            234 In Hammond (1997) 2 Qd R 195, the Court said at 199-200:
            “The true relevance of drug addiction as a factor contributing to the commission of crime and its effect in the sentencing process has never been adequately explained. In our view it is a factor that may help an offender to the extent of showing that his or her descent into the crime in question was a secondary consequence of desperation produced by a human weakness rather than a primary choice. It may be inferred in many cases (and the present case is very much in point) that the offender would almost certainly never have become a thief but for loss of control of an ordered life through drug addiction. The proposition that the results of drug addiction are self inflicted is half true and therefore dangerous. The offender must of course be held responsible for his or her own actions. The drug addiction is not an excuse; but it is a factor that may tell the court that the real weakness of character is that of a drug addict rather than that of a robber. That may be by no means inconsequential. It is however a two edged factor; it may also tell the court that rehabilitation is going to be difficult.
                Just how these factors are to be applied to particular cases has not been spelled out, and it may be undesirable to attempt to do so. In our view the fact that drug addiction has led to criminal activity can properly lead to a wider understanding of the offender’s character and motivation for the crime. Depending on the circumstances of the particular case this may sometimes show the offender to be less deserving of condemnation than the primary facts, without more, might suggest.”
            235 Later, the Court said at 211:
            “Addiction to drugs at the time of the offence is not an excuse, but may tell the court that the primary weakness of character is that of an addict rather than that of a robber. Such a factor tends to be mentioned along with an unfortunate family life as an explanation of the offender’s descent into bad habits. When the original problem was not something for which the offender was responsible, it is natural to feel some sympathy for the offender though this must not obscure the offender’s direct responsibility for the ultimate criminal act he or she chose to do.”
            236 These observations were relied upon by the respondents as providing some support for the proposition that the fact of drug addiction may properly lead to the culpability of the offender being regarded as less deserving of condemnation than would otherwise be the case. However, it is clear that the Court in Hammond did not question the widely accepted principle that drug addiction is not an excuse, and emphasised that the relevance of drug dependence will depend upon the circumstances of the particular case. The decision does not stand as authority for the general proposition advanced.
            237 Nolan (1998) VSCA 135 also provides only limited assistance to the respondents. Although Buchanan JA referred to drug addiction as a “mitigating” circumstance, the other member of the majority, Butt JA spoke of it as an “explanation”. It was an extreme case, where the impulsiveness of the offence, rather than the mere fact of drug addiction, took the matter outside the general rule recognised in Halewyn.
            238 Douglas (1995) 56 FCR 465 similarly was an extreme case. Von Doussa, Higgins and Nicholson JJ there accepted that, while it is not a mitigating factor that a person commits a crime to feed a drug addiction, nevertheless there will be extreme cases where that principle will not apply. The moral and legal fault for the acquisition of the addiction can properly be considered, for example where the offender becomes addicted at an age when the “degree of judgment” open to him or her is limited.
            239 The submission advanced essentially turned upon the proposition that, within the context of the national concern that currently exists in relation to the scourge of drug dependency, and its effects not only upon individual users and their families, but also upon the community as a whole, some greater flexibility should be exercised in sentencing those whose criminality is substantially linked to an addiction.
            240 In support of this submission reference was made to the views expressed by some academics and researchers that imprisonment provides no deterrent for this class of offender, and may tend only to exacerbate the problem.
            241 Additionally, reference was made to the need to look for new solutions or creative responses for what has proved to be an intractable pattern. Illustrations cited as either having been adopted or under review in this area include the Drug Court, expanded access to rehabilitation programmes, heroin trials, safe injecting rooms, and new forms of chemical therapy in the place of, or as an alternative to methadone (naltrexone, acamprosate, buprenorphine, slow release oral morphine and so on.)
            242 Within that context it was submitted that the bare statement that drug addiction is “not an excuse,” or is “not a factor in mitigation” reflects an approach by the courts which is too rigid. On the contrary, it was submitted a more constructive approach should now be adopted which gives greater recognition to the cravings of drug dependent offender who is forced through the absence of suitable alternatives, to resort to crime to feed a habit.
            243 Reference to the observation of Mildren J in Lewfatt (1993) 66 A Crim R 451, concerning the imprisonment of the typical offender with which the guideline judgment deals, underlies, at least in part this submission. His Honour there said:
            “The prisoner was young, and had undergone significant changes in her outlook. Her prior convictions were drug and alcohol related. She was now determined to stay off drugs and alcohol, and to pursue her ambitions for a career, and accept responsibility as a parent. Her history strongly suggested that she had gone through a difficult period of immaturity, and was now ready to rehabilitate herself. In those circumstances a period of actual imprisonment could well be counterproductive to this process. If one of the main purposes of punishment is to protect society, society’s interests are best served by a sentencing disposition which promotes the rehabilitation of the prisoner, rather than a disposition which may have the opposite effect.”
            244 To this statement can be added the observation of the majority of the High Court in Veen No.2 (1988) 164 CLR 465 at 476:
            The purposes of criminal punishment are various: protection of society, deterrence of the offender and of others who might be tempted to offend, retribution and reform. The purposes overlap and none of them can be considered in isolation from the others when determining what is an appropriate sentence in a particular case. They are guideposts to the appropriate sentence but sometimes they point in different directions.”
            245 In advancing the proposition that drug dependence should be recognised as a significant factor of mitigation, justifying a lesser sentence than might otherwise be appropriate, the respondents sought to draw upon material questioning the extent to which such dependence is a matter of full and free choice. An affidavit was tendered, for this purpose, sworn by Dr. Jurd, a psychiatrist specialising in the treatment of drug offenders, in which he deposed that:
            “drug addiction has a well documented neurobiological or physiological basis.”
            246 In addition he said that the:
            “predisposition to addiction has a substantial physiological (genetic) component. Thirty years of study of the neurobiological basis of alcoholism has revealed time and again that having a positive family history is among the most powerful predeterminants of alcoholism:
                (a) identical twins are more likely than fraternal twins to both suffer from alcoholism.
                (b) adopted away children of alcoholic parents, even if they have never lived with their parents, are more likely to themselves become alcoholic than other adoptees.
                (c) the risk of developing alcoholism increases depending on how many alcoholic relations one has, whether or not you live with them.
                (d) a series of molecular biological experiments have shown that at least one gene is over-represented among alcohol dependent people.
                There is every reason to extrapolate this information to other drugs. Similar studies have been done with drug dependents producing very similar results.”
            247 He continued by noting that drug addiction affects the addicts’ behaviour in predictable ways, altering values and overwhelming other priorities as the addiction becomes more severe.
            248 The hypothesis that addiction has a genetic component or predisposition was not tested. Nor was the validity of the comparison with alcohol dependency. Whether this view be correct or not, I am not persuaded that there is an inevitable causal relationship or progression from drug addiction to the commission of criminal offences, or to an incremental increase in the seriousness of these offences.
            249 In many instances, as studies tend to show, that may well be he case: Dobinson et al Drugs and Crime: A Survey of New South Wales Property Offenders 1984. New South Wales Bureau of Crime Statistics and Research, 56, 57, 62 and 64; Dobinson Making Sense of the Heroin-Crime link (1882) 22 ANZJ Crim, 269; Maher et al Running the Risks: Heroin, Health and Harm in South West Sydney National Drug & Alcohol Research Centre University of New South Wales Monograph No 38, 51 and 62; Hall, Methadone Maintenance Treatment as a Crime Control Measure Bureau of Crime Statistics and Research (1996) 2; Hall The Role of Legal Coercion in the Treatment of Offenders with Alcohol and Heroin Problems (1997) 30 ANZJ Crim 105; Lipton The Effectiveness of Treatment for Drug Abusers under Criminal Justice Supervision (1995) National Institute of Justice, 4.
            250 However, many offenders commence their criminal careers before turning to drugs, and not all persons who become addicted to drugs commit armed robberies or even serious property offences.
            251 Moreover, there are many factors other than drug dependency, such as multiple social disadvantage or an attachment to a criminal sub culture, that contribute to criminal behaviour; and it would, accordingly, be too simplistic to generalise that drug dependency is inevitably the precursor to or cause of criminality or of recidivism.
            252 Nor am I persuaded of the appropriateness of the suggested analogy between drug addiction and mental abnormality, in respect of which the element of general deterrence is often given less weight: Veen (No.2) at 476-477, Scognamiglio (1991) 56 A Crim R 81; Tsaris (1996) 1 VR 398; Letteri (Court of Criminal Appeal New South Wales 18 March 1993) and Engert (1995) 84 A Crim R 67.
            253 The relevant principle as stated in Letteri by Badgery-Parker J, in a passage adopted by Gleeson CJ in Engert is as follows:
            “… that whereas general deterrence is a relevant consideration in every sentencing exercise, it is a consideration to which less weight should be given in the case of an offender suffering from a mental disorder or severe intellectual handicap. In an extreme case, the proper application of this principle may produce the result that considerations of general deterrence are totally outweighed by other factors. In every case it is a matter of balancing the relevant factors in a manner no different from that which is involved in every sentencing exercise.”
            254 The reason for this approach lies in the circumstance that the community will readily understand that the offender who suffers from a mental disorder or abnormality is less in control of his or her cognitive facilities or emotional restraints, and in some instances lacks the ability to make reasoned or ordered judgments. Almost invariably there is a limited appreciation of the wrongfulness of the act, or of its moral culpability, which although falling short of avoiding criminal responsibility does justify special consideration upon sentencing. Moreover, such a condition is inherent and its presence does not depend upon any element of choice.
            255 The case of the drug dependent offender is very different in so far as the community is likely to see such a person as a perfectly appropriate medium to stand as an example to others.
            256 These differences were implicitly recognised in Terizakis (1986) 41 SASR 252 where O’Loughlin J, with whose reasons Cox J agreed, said at 256:
            “… heroin addiction cannot, of itself, justify a sentencing judge reducing a penalty, which is otherwise appropriate to the offence that has been committed and to the circumstances in which the offence was committed. One cannot help but have immense sympathy for a person who is addicted to heroin; and it would be callous to adopt an attitude that, because it is self-inflicted, the addict is not worthy of sympathy or concern. But having said that, it unfortunately remains a fact of life that all too many armed hold-ups are these days committed by persons who, in one way or another, are dependent upon drugs and are seeking money to meet the costs of that dependency. Despite the cravings caused by the addition and despite irrationality that is so often a consequence of the addiction, the majority of drug addicted persons still retain, in varying measures, the ability to distinguish between right and wrong and the ability to comprehend that the consequence of armed robbery will be a lengthy term of imprisonment. To give way, out of sympathy and concern, by imposing lesser terms of imprisonment than would otherwise be imposed, might appear to be an act of humanity, but it will do nothing to stamp out the drug trade and, worse, it might even give some indirect encouragement to would-be offenders.”
            257 The comparison is also imperfect in so far as it overlooks the original element of choice every person has, initially whether or not to experiment with drugs, and thereafter to continue with their use. Even if some persons have a genetic predisposition or vulnerability to addiction, they are not, in my view, entitled to claim any favours in that regard. The addictive quality of drugs and the potential of a slide into a degraded or criminal lifestyle are so well known as to invite a free choice. Those who choose to use drugs must, in my view, accept the consequences.
            258 Moreover, as Gleeson CJ observed in at 71:
            “…the question of the relationship, if any, between the mental disorder and the commission of the offence, goes to circumstances of the individual case to be taken into account in the application of the relevant principles. The existence of such a causal relationship in a particular case does not automatically produce the result that the offender will receive a lesser sentence, any more than the absence of such a causal connection produces the automatic result that an offender will not receive a lesser sentence in a particular case. For example, the existence of a causal connection between the mental disorder and the offence might reduce the importance of general deterrence, and increase the importance of particular deterrence or of the need to protect the public. By the same token, there may be a case in which there is an absence of connection between the mental disorder and the commission of the offence for which a person is being sentenced, but the mental disorder may be very important to considerations of rehabilitation, or the need for treatment outside the prison system.”
            259 By the same process of reasoning the existence of a causal relationship between a drug habit and the commission of an offence should not, as a matter of general principle, automatically lead to a lesser offence.
            260 The respondents also relied upon the proposition that there was little evidence that the punishment of offenders has any general deterrent effect, whether for offenders at large or drug dependent persons in particular. It may be recognised that while there are different views about this, some studies have been interpreted as showing that the perceived severity of a sentence, as distinct from the certainty of detection and arrest, does not of itself provide a deterrent effect: Anderson et al “Formal and Informal Sanctions: A comparison of deterrent effects” (1997) Social Problems 25: 103-117; Paternoster et al “Perceived Risk and Social Control: Do Sanctions really deter?” (1983) Law and Society Review 17: 467-479; Baron et al “Deterrence and Homeless Street Youths (1998) Canadian Journal of Criminology 27-60; James Austin et al “Does Imprisonment Reduce Crime? A Critique of ‘Voodoo’ Criminology” (1993) National Council on Crime and Delinquency. 10, 22.
            261 In the case of drug dependent persons contemplating the commission of an armed robbery, in particular, it was suggested, that the prospect of imprisonment provides little deterrent, in that:
            · their addiction causes them to give high priority to drug seeking behaviour, sometimes to the exclusion of all else;
            · they are accustomed to engaging in high risk activity: Maher et al “Running the Risks: Heroin, Health and Harm in South West Sydney” National Drug and Alcohol Research Centre University of NSW NDARC Monograph no. 38 at 26, l20, 125; Hall Methadone Maintenance Treatment as a Crime Control Measure (1996) Bureau of Crime Statistics and Research, 2.
            · the clearance rate for armed robbery is less than 1 in 5 (Chilvers: New South Wales Recorded Crime Statistics 1998 at 30-32;
            · there is no realistic prospect of law enforcement activity reducing the size of the drug addicted population or of it reducing the street level price of drugs (thereby alleviating the financial pressure to resort to crime to support a habit) Maher supra at 126) and
            · despite the steady increases in sentences, following the introduction of the Sentencing Act 1989, and also amendment of the Crimes Act 1900 to increase the maximum available penalty, in respect of many offences, the crime rate has not fallen.
            262 It was additionally submitted, there was little evidence that lengthy periods of imprisonment have any substantial rehabilitative consequences for the drug dependent offender. To the contrary, the argument ran, and as recidivism rates suggest, incarceration may be counterproductive, initiating or encouraging the offender into criminal behaviour, and breaking social and familial bonds which militate against such behaviour.
            263 Alternative sentencing options, it was submitted, were more likely to reduce drug dependency and the consequent incidence of criminality and of recidivism: Hall: “Methadone Maintenance Treatment as a Crime Control Measure” (June 1996) Contemporary Issues in Crime and Justice, Bureau of Crime Statistics and Research 3, and 6-7; Hall: “The Role of Legal Coercion in the Treatment of Offenders with Alcohol and Heroin Problems” (1997) Vol 30 ANZJ Crim 113; Lipton: “The Effectiveness of Treatment for Drug Abusers Under Criminal Justice Supervision” (1995) National Institute of Justice 51-53; Murphy: “Drug Courts: An Effective Efficient Weapon in the War on Drugs” (1997) Illinois Bar Journal Vol 85 487; Tauber: Drug Courts: a Judicial Manual California Centre for Judicial Education and Research (1994) 1-2 and 9-10; and Belenko “Research on Drug Courts: A Critical Review” (1998) National Drug Court Institute Review Vol 1 Issue 1, 21-23, 29 and 35.
            264 The views expressed in these studies concerning the deterrent value of imprisonment, and the possible exposure of prison inmates to influences that may increase their prospects of recidivism, are not confined to drug dependent offenders. They are of common application, particularly in the case of first offenders, for whom the slamming prison door principle remains apposite, and also in the case of those who are yet to offend, for whom the prospect of imprisonment has a particular relevance.
            265 I am not prepared to advocate any departure from the long accepted wisdom that imprisonment does have a personal and general deterrent effect. It is a notion deeply entrenched in the criminal law, and it has the imprimatur of the legislature which has prescribed significant maximum penalties for the offence under consideration, as well as for other offences involving serious criminality.
            266 Moreover, it cannot necessarily be assumed from the fact that increases in sentences have not been accompanied by any noticeable drop in crime rates, that they lack deterrent effect. In the absence of any control, it cannot be known whether that crime rates would have been higher had sentences not been increased.
            267 The risk of exposure to undesirable influences within a corrective environment, similarly cannot be used as a justification for abandoning, or even limiting imprisonment, as a general response to criminal conduct. There remains for every offender a choice between reform and recidivism, and the problem is better addressed by the development of adequate programs and rehabilitation options within the prison environment, than it is by a significant change in sentencing policy.
            268 While it may be accepted that the rehabilitative aspect of sentencing is not an unimportant consideration: Osenkowski (1982) 30 SASR 21; Douglas (Court of Criminal Appeal New South Wales 4 March 1997); Eastway (Court of Criminal Appeal New South Wales 19 May 1992); Fabian (1993) 64 A Crim R 365; and Halewyn, I am not persuaded that the offender, whose criminality was directly influenced by drug addiction, should be placed into a special category for sentencing purposes.
            269 The objective of rehabilitation is but one aspect of sentencing. It needs to be taken into account along with the objectives of retribution and deterrence.
            270 It is more appropriate, in my view, to have regard to these considerations, upon an individual case basis, in which it is recognised that the offence of armed robbery remains a very serious crime, which, save in exceptional circumstances, calls for full time imprisonment. That is not to say that, in an individual case, the fact of drug dependency of the person standing for sentence may not remain a relevant fact reflecting on the objective circumstances of the offence, and/or the subjective circumstances of the offender, and particularly on the relativity of the minimum and additional terms: Bugmy (1990) 169 CLR 525 at 537; and B (1993) 68 A Crim R 547.
            271 This approach accords with the caution offered by Gleeson CJ in Engert at 68:
            “It is … erroneous in principle to approach the law of sentencing as though automatic consequences follow from the presence or absence of particular factual circumstances. In every case, what is called for is the making of a discretionary decision in the light of the circumstances of the individual case, and in the light of the purposes to be served by the sentencing exercise.”
            272 It also avoids the problems associated with establishing categories according to the offender’s need for money (e.g. to feed a drug addiction, to support a gambling habit, to pay off business or family debts, to establish a business, etc) which would lend themselves to abuse and to moral judgments in respect of which minds may legitimately differ.
            273 In my view the relevant principles are as follows:
                (a) the need to acquire funds to support a drug habit, even a severe habit, is not an excuse to commit an armed robbery or any similar offence, and of itself is not a matter of mitigation;
                (b) however the fact that an offence is motivated by such a need may be taken into account as a factor relevant to the objective criminality of the offence in so far as it may throw light on matters such as:
                    (i) the impulsivity of the offence and the extent of any planning for it; (cf Bouchard (1996) 84 A Crim R 499 at 501-2); and Nolan (1988) VSCA 135 (2 December 1998);
                    (ii) the existence or non existence of any alternative reason that may have operated in aggravation of the offence, eg that it was motivated to fund some other serious criminal venture or to support a campaign of terrorism;
                    (iii) the state of mind or capacity of the offender to exercise judgment, eg if he or she was in the grips of an extreme state of withdrawal of the kind that may have led to a frank disorder of thought processes or to the act being other than a willed act;
                (c) It may also be relevant as a subjective circumstance, in so far as the origin or extent of the addiction, and any attempts to overcome it, might:
                (i) impact upon the prospects of recidivism/rehabilitation, in which respect it may on occasions prove to be a two-edged sword (eg Lewis Court of Criminal Appeal New South Wales 1 July 1992);
                (ii) suggest that the addiction was not a matter of personal choice but was attributable to some other event for which the offender was not primarily responsible, for example where it arose as the result of the medical prescription of potentially addictive drugs following injury, illness, or surgery (cf Hodge Court of Criminal Appeal New South Wales 2 November 1993; and Talbot); or where it occurred at a very young age, or in a person whose mental or intellectual capacity was impaired, so that their ability to exercise appropriate judgment or choice was incomplete;
                (iii) justify special consideration in the case of offenders judged to be at the “cross roads”: Osenkowski(19882) 5 A Crim R 394.
            274 To go further, and to accept the fact of drug addiction as a mitigating factor generally, would not be justified in principle. Moreover, it would involve an exercise in irresponsibility on the part of the Court, if it were understood as a message that committing the crime of armed robbery to feed a drug habit is less deserving of censure than would otherwise be the case.
            275 The legislature has, by the heavy maximum penalty prescribed for armed robbery, spoken clearly in relation to this offence. Drug dependent persons should not be encouraged, as a class, to think that they are free to engage in serious criminal conduct of whatever kind with impunity, or with any hope of favourable treatment because they are able to show that they needed money through their addiction.
            276 In summary, I see no reason to depart from the planks of punishment, retribution and rehabilitation that underlie the sentencing process generally, and that permit of individualised sentencing by reference to the objective and subjective circumstances of each case.
            277 I do not suggest by these observations that greater attention and resources should not be dedicated to drug rehabilitation generally, both within and outside the prison environment. Clearly it is desirable that alternative approaches to the problem be trialed, it being far preferable that potential criminality be headed off than punished after the event.
            **********

            IN THE COURT OF

            CRIMINAL APPEAL
                                60559/98
                                60558/98
                                60561/98
                                60746/98
                                60596/98
                                60595/98
                                60511/98
                                            SPIGELMAN CJ
                                            WOOD CJ at CL
                                            NEWMAN J
                                            HULME J
                                            SIMPSON J
                                        Wednesday 12 May 1999

                      REGINA v Paul Anthony HENRY
                      REGINA v Stephen Anthony BARBER
                      REGINA v Hoai Vinh TRAN
                      REGINA v Troy David SILVER
                      REGINA v Theo TSOUKATOS
                      REGINA v Bill KYROGLOU
                      REGINA v John David Jenkins

              JUDGMENT

              278 NEWMAN J: I have read the reasons of Spigelman CJ relating to the implementation of guide lines in respect of sentencing in relation to the crime of armed robbery and whether drug addiction can be taken into account by way of mitigation. I agree with those reasons. I also agree with Wood CJ at CL.
              **********

              IN THE COURT OF
              CRIMINAL APPEAL
                                  60559/98
                                  60558/98
                                  60561/98
                                  60746/98
                                  60596/98
                                  60595/98
                                  60511/98
                SPIGELMAN CJ
                                            WOOD CJ AT CL
                                            NEWMAN J
                                            HULME J
                SIMPSON J
                Wednesday 12 May 1999
                REGINA v Paul Anthony HENRY
                REGINA v Stephen Anthony BARBER
                REGINA v Hoai Vinh TRAN
                REGINA v Troy David SILVER
                REGINA v Theo TSOUKATOS
                REGINA v Bill KYROGLOU
                REGINA v John David Jenkins

                GUIDELINE JUDGMENT

                HULME J:
                Guideline Judgment
                279 I have read in draft the Reasons of Spigelman CJ. I agree with his Honour’s conclusion that a guideline judgement is required and generally with His Honour’s reasons for that conclusion. However, on that topic there are some additional remarks I would make.
                280 Firstly, I am satisfied that the incidence of armed robbery has had an impact on society by changing patterns of behaviour out of a sense of apprehension. The increasing use over recent, perhaps 10 or so, years of time delay locks, video cameras and of the practice that safes can only be opened by employees of companies such as Brambles with their armoured vans, I have no doubt, is a reaction to this type of conduct.
                281 Secondly, the evidence adduced by the Crown satisfies me that there has been a substantial increase in the incidence of armed robbery. In this regard four sources of data were presented to the court: Victim Surveys, Court Statistics, Statistics kept by the NSW Bureau of Crime Statistics and Research, and information in the Report on Government Services 1999 released by the Steering Committee for the Review of Commonwealth/State Service Provision.
                Victim Survey Data
                282 The Victim Survey data was contained in annexure B14 to the Crown’s original submissions. This was a May 1998 “Crime and Justice” bulletin published by the NSW Bureau of Crime Statistics and Research. The article in that bulletin deals with “robbery” generically and records that it is based on what was said to be “a representative sample of about 6,500 NSW households involving some 13,000 individuals”. It notes however that there was some “confusion about the definition of robbery (resulting) in a significant over estimate of the incidence of robbery in NSW between 1990 and 1995”. A change in the definition of robbery was then used in the survey with the result that data collected in 1996 and 1997 is not comparable with that collected earlier.
                283 The article purports to record victimisation rates for robbery between 1990 and 1997. The figures for 1996 and 1997 are recorded as 0.4% and 0.3% respectively. The figures for other years have to be derived from a rather small chart and, although precision is impossible, seem to be as set out below. The direction of movement from year to year is a correct reflection of the chart.
                1990 1.7
                    1991 1.6
                    1992 1.1
                    1993 1.3
                    1994 1.2
                    1995 1.8
                284 The article also recorded that in 1997 over 65% of the respondents to the survey who had been a victim of a robbery indicated that there was no weapon used and less than 3% of the robbery victims indicated that a firearm had been used.
                Criminal Court Statistics
                285 These were extracted from the relevant pages of seven annual publications produced by the NSW Bureau of Crime Statistics and Research and were:-


                YearNumber of Persons ChargedNumber of Charges Finalised
                19918531306
                19929701481
                19938341251
                19947681051
                1995681943
                1996708976
                19977841128
                286
                In its submissions the Crown said that “the publication records the number of people charged separated by the principal offence committed by each person (i.e. the most serious offence) and the total number of charges finalised”. However, reference to page 54 of the full 1996 publication suggests that this is not so and that, whatever other counts a person may have faced, so long as one count was robbery that person is included within the “number of persons charged”. I have not found a clear explanation of the “number of charges finalised”. It is not apparent whether it reflects the number of counts of robbery dealt with or the number of counts dealt with in the case of offenders who faced one count of robbery.
                NSW Bureau of Crime Statistics and Research
                287 The third set of statistics are from publications of the NSW Bureau of Crime Statistics and Research recording information supplied by the police. The Court was informed that methods of recording and correlation have changed over the years and that it is not possible to directly compare figures prepared under one system with those prepared under another. Although one cannot directly compare the figures in one of the periods referred to below with the figures in another, it is possible to compare figures within a period with other figures in the same period.
                1982/83 - 1989/1990
                288 Annexure B11 to the Crown’s original submissions was an extract from the NSW Recorded Crime Statistics 1989. It shows the following rate of recorded offences per 100,000 population.
                Offence82/8383/8484/8585/8686/8787/8888/8989/90
                Robbery with firearm32.522.120.315.516.417.816.514.4
                Robbery with a weapon not a firearm12.09.812.515.416.719.519.722
                Total rows 1 & 244.631.932.830.933.137.236.236.4
                Robbery - no weapon37.637.436.030.137.338.741.644.2
                Total rows 3 & 482.269.368.861.070.475.977.880.6

                Note: The total of rows 3 and 4 has been inserted by me.

                289 So far as armed robbery is concerned, the figure for 1982/83 is so high it looks to be an aberration. The figures show an increase of about 13% between 1983/84 and 1989/90 or between the three year period 83/84 - 85/86 on the one hand and the three year period 87/88 - 89/90 on the other. The figures for “robbery - no weapon” are suggestive - I put the matter no higher - of an increasing trend.
                1991-1993
                290 The information for 1991/93 is much briefer. It comes from the Bureau publication NSW Recorded Crime Statistics 1993 and is contained in Annexures B9-B10 to the Crown’s original submissions. The commentary records that for robbery overall “there was no statistically significant upward or downward trend over the 3 year period to December 1993”. However, it is recorded that “there was a statistically significant downward trend in robbery with a firearm … a 30.7% decrease”.
                April 1994-December 1995
                291 The information covering the period April 1994 to December 1995 is again brief. The Court was informed that it came from an August 1996 Crime and Justice Bulletin published by the Bureau. The extract forms Annexure B8 to the Crown’s original submissions. The commentary records:-
                “From the first 9 months to the last 9 months of the period April 1994 to December 1995, robbery without a weapon increased by 7.3% and robbery with a weapon other than a firearm went up by 27.3%. The recent trend for robbery with a firearm has been neither upward or downward.”
                292 It is not possible from the evidence to say whether this information relates to the absolute number of offences or is related to population. However, one can be confident that the population of New South Wales did not increase at the rate of 27.3% in this period.
                1995-1998
                293 Information for 1995-97 comes from the Bureau publication NSW Recorded Crime Statistics 1997. It is contained inter alia in Annexures B4-B6 of the Crown’s original submissions. This is extended to 1998 in an extract from the 1998 publication which formed Annexure C to the further Crown submissions. Apart from a decrease between 1997 to 1998 in the case of robbery with a firearm and an (insignificant) decrease in the case of robbery without a weapon in the same period, these figures show a substantial increase in offences. The figures per 100,000 population are:-
                Offence1995199619971998
                Robbery with firearm11.113.317.5/18.214.4
                Robbery with weapon not firearm2430.553.4/53.969.2
                Total rows 1 & 235.143.870.9/72.183.6
                Robbery without a weapon73.979.8102.2/103.4101.9
                Total rows 3 & 4109123.6173.1/175.5185.5

                Notes: Totals have again been inserted by me.
                    Annexures B4-B6 on the one hand and Annexure C on the other state different figures for 1997.
                294 At face value these figures indicate that between 1995 on the one hand and 1997 and 1998 on the other there has been an increase of over 100% in the incidence of armed robbery, more so where the “arm” has not been a firearm.
                295 Some documents show the absolute number of offences committed during some of the periods referred to above. Annexure B12 to the original Crown submissions and Annexures B4-B6 and C to the further Crown submissions reveal:-

                Offence82/8383/841995199619971998
                Robbery no weapon20042009392643275633/56925511
                Robbery with firearm17331187597732 962/1001 821
                Robbery with weapon not firearm 641 529128916562951/29783861

                Note: Annexures B4-B6 on the one hand and Annexure C on the other state different figures for 1997.

                296 I do not forget the remarks made above about the difficulty of comparing figures for one period with figures in another . However, these figures show that the number of offences of robbery without a weapon has at all times been a substantial proportion of the total, and during 1995-1998 significantly exceeded the total of the number of offences involving a weapon whether or not a firearm. It follows from these figures that the Victim Survey Data and Criminal Court statistics which refer to robberies generically are of no assistance in making any judgment on the question of whether there has been a change in the incidence of armed robbery.
                Report on Government Services 1999
                297 This document is Annexure A to the Crown’s supplementary submissions. It records that on the basis of crimes reported to police the incidence of armed and unarmed robbery per 100,000 of population in Australia and its various states and territories during 1993 to 1997 was as follows:-

                Armed Robbery
                YearNSWVICQLDWASATASACTNTAUST
                199338.622.230.328.333.59.618.18.330.2
                199436.717.327.034.331.813.321.38.228.3
                199562.316.826.638.724.812.925.79.836.7
                199648.618.627.254.823.110.131.814.334.1
                199779.325.736.658.925.39.536.212.248.6
                Unarmed Robbery
                YearNSWVICQLDWASATASACTNTAUST
                199361.520.532.029.782.912.920.419.542.3
                199484.819.033.537.371.318.021.922.850.0
                199594.521.132.942.274.413.527.033.454.5
                199693.623.436.542.367.420.035.145.155.3
                1997121.027.835.059.357.422.037.426.766.1


                298 The combined figures for New South Wales, per 100,000 population are set out in the table below. For the purposes of comparison, I have set out in a third column the comparable Bureau of Crime Statistics figures where available.
                1993 101.1
                    1994 121.5
                    1995 156.8 109
                    1996 142.2 123.6
                    1997 200.3 173.1/175.5
                299 Albeit there are some “dips”, the figures for New South Wales show a decided upward trend. Except for South Australia and, in the case of armed robbery, Tasmania, the other states and territories also show an upward trend in both classifications. Australia wide, including New South Wales, there was an increase in each category between 1993 and 1997 of over 50%.
                Conclusion
                300 The inconsistency between the above sets of figures raises the question whether any can be relied upon. At first blush each has doubt thrown on it by one or more of the others.
                301 The change in definition used in the compilation of the Victim Surveys does not inspire confidence. The limited extent of the sample leads to some doubt as to its validity. The proportion of unarmed robberies recorded in that survey of over 65% is not substantially different from the proportion of these recorded in the Bureau of Statistics figures for 1997 (59%), but the proportion of robberies with a firearm - less than 3% - is radically different from the Bureau of Statistics figures over many years. The Bureau information for the periods 1982/83 - 1989/90, 1991 - 1993, and 1995 - 1998 has consistently shown such offences to be a decreasing proportion of robberies as a group but the lowest proportions ever have been in 1997 (10.1 - 10.3%) and 1998 (7.7%). Given the consistency in these Bureau figures and the period involved, the proper conclusion to draw in respect of the survey figures relating to robberies with a firearm is that they are unreliable. That throws considerable doubt on the validity of the survey.
                302 The only possible support any of the other data provide for the Survey figures is in the Criminal Court statistics which, for the period 1992 to 1995 echo to some extent the decreasing trend in the Survey figures for all robberies for the period 1990 to 1994, some time lag in the former figures being expected.
                303 In that the Criminal Court statistics do not take into account offences not leading to a person being charged and it is not clear that they even detail the number of offences the subject of a charge, they have obvious limitations. Neither do they indicate the extent to which at the beginning and end of the periods under consideration, there may have been a backlog of cases. The Criminal Court statistics are also in absolute numbers and I confess to some surprise at the suggestion implicit in those figures that robbery has actually decreased since 1991 or 1992, despite the notorious increase in drug taking which has occurred. On the other hand, the information contained in the Report on Government Services does show decreases in some states over some periods. An example is Victoria between 1993 and 1996.
                304 The Bureau of Crime Statistics information does have a deal of consistency about it over an extended period. In saying that I do not of course suggest that there have not been departures from trend lines and again I do not forget that direct comparison of data between periods is not possible. I have referred to the relative decrease in robberies with a firearm over three of the four periods covered by this information. There is a similar consistency, albeit in the other direction, in the data relating to robbery with a weapon other than a firearm within the periods 1982/83 - 1989/90 (except for 1982/83 itself), April 1994 - December 1995, and 1995 - 1998 and in the data relating to robbery without a weapon within the periods 1982/83 - 1989/90 (except for the first 3 of these 8 periods), April 1994 - December 1995, and 1995 - 1998. In that the information for 1991 - 1993 records that there was no statistically significant trend overall but a significant downtrend in robbery with a firearm, during that period, it indicates that there was an increase in robbery without a weapon or robbery with a weapon other than a firearm, or both, during the period. Given the length of time between 1982/83 and 1998, the exceptions I have noted do not invalidate the conclusions I have drawn as to the consistency of the information.
                305 The difference between the information from the Bureau of Crime Statistics and that in the Report of Government Services suggests that it could not have come from the same source. There are dips in 1994 and 1996 in the armed robbery figures and these are, of course 2 out of the 5 years referred to. On the other hand the figure for 1996 is appreciably higher than for 1994 and the other three years all show an increase. The unarmed robbery figures all tend in the one direction apart from those for 1996 which are only marginally less than for 1995. Considered overall, it seems to me that the information in the Report tends in the same direction as that from the Bureau, albeit for the much shorter period of 1993 to 1997. Thus although the Report figures cast doubt on those from the Bureau, both sets of data support the conclusion that there has been a significant increase in the incidence of offending in New South Wales.
                306 Absent other evidence, I would have been disposed to regard the Bureau of Crime Statistics as a reliable source of information in respect to statistics falling within its province such as those set out above. Having considered them, the Victim Survey data and Criminal Court statistics to which I have referred do not cause me to doubt the reliability of the Bureau statistics. If I had to choose between them, I would be disposed to regard the statistics from the Bureau as more likely to be reliable than that from the Report on Government Services but because of the limited purpose to which it is urged the statistical data should be put, I do not need to make that choice. The information from both the Bureau and the Report demonstrate that there has been a very substantial increase in the incidence of both robbery and armed robbery. The statistics from these two sources are sufficiently consistent and the periods covered by them sufficiently long that, subject to any statutory or discretionary restraints, the Court should approach the issues before it on that basis.
                307 At the risk of repetition, the information from the Report indicates that in New South Wales between 1993 and 1997 there was an increase in armed robbery incidents per 100,000 population from 38.6 to 79.3 - 105%. If the average of 1996 and 1997 is compared with the average of 1993 and 1994, the increase is 70%. The increase in unarmed robbery and robbery overall is about 100%.
                308 The information from the Bureau of Crime Statistics indicates-
                Between 1983/84 and 1989/90, or between the three year period 83/84 - 85/86 on the one hand and the three year period 87/88 - 89/90 on the other, the incidence of armed robbery increased by about 13%.
                    In the period 1991 to 1993, in robbery overall there was no statistically significant trend albeit robbery with a firearm decreased by about 30.7%.
                    From the first to the last 9 months of the period April 1994 to December 1995, robbery with a weapon other than a firearm went up by 27.3%. The trend for robbery with a firearm was neither upward or downward.
                    Between 1995 and 1998 there was an increase in armed robbery incidents per 100,000 population from 35.1 to 83.6 - 138%. The percentage increase is 188% if robbery with a weapon not a firearm is the category considered.
                309 The size of the increase between 1995 and 1998 is staggering. Given that the period covered by the figures for those years is only four years and the change is so much greater than that apparent for earlier periods, I am disinclined to rely on that data as showing reliably in quantitative terms, a long term trend. However, when regard is had to the fact that the New South Wales statistics for three of the four periods mentioned show a substantial increase in the incidence of armed robbery during those periods, and the Government Services report shows the same for the period it covers, both for New South Wales and Australia generally, I am of the view that one should accept that the offence has become far more common in this State, not merely in absolute terms but also per head of population. It must be remembered that insofar as the percentage increases to which reference has been made are accepted, they need to be compounded if one is considering the change since about 1983.
                310 I would add that, given the increasing usage of heroin and other drugs in the community and the need of most addicts to fund their addiction by crime, the conclusion that there has been a substantial increase in the incidence of armed robbery is not surprising.
                311 It was not suggested on behalf of the respondents to the appeals that resort to statistical material of the above kind was other than legitimate and no other inconsistent data was advanced. Certainly, the reliability of the data was criticised but I have dealt with that topic. The Court’s attention was directed to the victimisation rates for 1996 and 1997 of 0.4 and 0.3% respectively and it was submitted that these showed that armed robbery was a reasonably rare offence. However these figures indicate that the average chance of being robbed during a 50 year adult life span is between 15 and 20%. It is not obvious that that prospect should be regarded with equanimity.
                312 The increased incidence in armed robbery, particularly where the weapon is not a firearm, justifies a review of the level of sentencing that has applied hitherto and thus a guideline judgment.
                Evidence as to the Impact on Victims
                313 Turning to another topic, I take the view that the evidence as to the impact of robbery on victims generally is a matter to which the Court may have regard either as a matter of judicial notice or as an “other matter” within s29(1) of the Criminal Procedure Act and a matter the Court is entitled to take into account on the question of sentence in the individual cases. It is of the same nature as the material the Court was referring to in R v Broxam (unreported, CCA, 3 April 1986):-
                “The prevalence of the offence in respect of which the applicant was sentenced is a matter of grave social concern… It is now, well recognised that the victims of such armed robberies or attempted armed robberies may suffer devastating psychological damage consequent upon the trauma necessarily involved with the commission of such offences. In these circumstances sentencing judges would be recreant to the trust which the community places in them, if they were not to impose sentences consistent with the seriousness and prevalence of such offences.”
                314 The passage was cited with approval in R v Bell (unreported, CCA, 11/5/93) and R v Stefanovski (unreported, CCA, 9 June 1994).
                315 That material is relevant to sentencing standards generally and therefore to the individual cases here. Of course, any evidence as to the impact on the particular victims is also relevant whether that impact be negligible or major. But if there is no particular evidence, it does not seem to me that the Court must proceed on the basis that there has been no impact or that it is not entitled to inform itself as to the possible impact.
                316 The full terms of Section 29A of the Criminal Procedure Act 1986 are set forth in paragraph 56 of the judgment of the Chief Justice. Sub-section (1) allows the Court to take into account “evidence or other matters”. Sub-section (2) provides that the “Court must not increase a sentence in any appeal by reason of, or in consideration of, any evidence that is used by the Court in giving a guideline judgment in the appeal and that was not given at the trial”. Normal principles of statutory construction lead to the conclusion that there is no prohibition in sub-section (2) in using “other matters” to increase a sentence. Within this expression would be the material referred to in Veen v R (No 2) (1987-88) 164 CLR 465 at 473 and that to which this Court had regard in R v Hollocoglu (1991) 29 NSWLR 67.
                The Guideline for New South Wales
                317 On the question of what the guideline for New South Wales should be, I agree with Spigelman CJ that the category of case to which His Honour referred is sufficiently common for the purposes of setting a guideline. However, in my view the guideline should be set at a level higher than that proposed by the Chief Justice. It should be set so as to include a total term of penal servitude of 5 years and, except in unusual circumstances, a minimum term of 3 years even when special circumstances within Section 5 of the Sentencing Act 1989 (NSW) are found. Absent special circumstances, the minimum term would be, as required by that section, not less than 3¾ years.
                318 A minimum term should be included in the guideline because inadequacy in the length of such terms is one of the deficiencies in past sentencing practice which is apparent in the review of the Judicial Commission sentencing statistics referred to in the judgment of the Chief Justice. Given the frequency with which special circumstances have been found - and there is nothing to suggest this is likely to change - and minimum terms of less than half the full term have been imposed - matters referred to in paragraph 107 of the Chief Justice’s reasons - it is likely that, absent the inclusion of a minimum term in the guideline, there will continue to be many sentences imposed for armed robbery where the total term is 4 years and the minimum term is 2 years or less. As guidelines, I regard the former figure as too low and the latter as wholly inadequate for a term which must “remain such as appropriately reflects the criminality involved” - Morrissey (unreported, CCA, 15 July 1994, cited with approval in R v McDonald (unreported, CCA, 12 October 1998).
                319 The statutory provisions relevant for present purposes are those contained in Sections 94 to 98 of the Crimes Act 1900 (NSW). Although actions other than robbery with or without aggravating features are encompassed within these Sections, it will be convenient in the discussion which follows to refer to the term rob and its derivatives as encompassing the all of the primary activities proscribed. The sections reveal a clear pattern of seriousness in the mind of the legislature.
                · Simple robbery carries a maximum punishment of penal servitude for 14 years - Section 94.
                · Robbery in circumstances of aggravation, or in company, or being armed with an offensive weapon, or accompanied by the malicious infliction of actual bodily harm carries a maximum punishment of penal servitude for 20 years - Sections 95 and 97(1).
                · Robbery accompanied by:-
                    (a)Wounding - Section 96, 98,
                        (b) The infliction of grievous bodily harm - Section 96, 98, or
                        (c) Being armed with a dangerous weapon - Section 97(2),
                        increases the maximum penalty to one of 25 years.
                320 (The inclusion by s4 of a dangerous weapon within the definition of “offensive “ weapon, if the definition applies throughout these provisions, does complicate the orderly pattern otherwise revealed. However, it does not affect the substance of the above picture.)
                321 Commonly the objective circumstances which may be involved in offences against this group of sections may be considered as falling within four areas, viz. the amount of money or the value of the items involved, the premeditation and degree of planning on the part of an offender, the offender’s actions towards his victim or victims and fourthly the impact of those actions on the victim or victims. It is appropriate that, to discourage such activity, the law impose high sanctions on those who participate in criminality for high stakes. Those whose criminality extends over a greater period and displays deliberateness and planning can, all other things being equal, anticipate the duration of their criminal intent and actions to be reflected in any sentence imposed.
                322 In considering the actions of an offender, it should be recognised that robbery is a denial of one of the elementary freedoms on which our society is based. Robbery involves the imposition of the will of the offender by force or the threat of force on the rights and ability of the victim to go about his own affairs in his own way. Some actions on the part of offenders, e.g. being in company, being armed, the malicious infliction of actual bodily harm, find express recognition in the statement of the offences. Others are also relevant. Thus conduct calculated to cause terror rather than merely the fear necessarily involved in an offence is an aggravating circumstance. Relevant also are the risks created by an offender - Readman 47 A Crim R 181 at 185, Lane (unreported, CCA, 3 November 1995). A consideration of the cases demonstrates that not all victims meekly acquiesce in a robber’s demands. Commonly a robbery occurs in circumstances charged with emotion and experience within and without the courts shows that there is but a fine line between a threat and a weapon on the one hand and injury or death on the other. Although an offender who crosses that line may lay him or herself open to further charges, it is important that the courts impose real sanctions to discourage the creation of the risk.
                323 Some impact on the victim is specifically referred to in the statutory provisions. But it is also important to bear in mind that it is an essential element in every robbery that the victim will be put in fear of an offender and of what that offender might do. After all it is the engendering of that fear which it is the offender’s intention and by which he is able to carry out his subjection of the victim or his will.
                324 In many cases the fear will be of imminent death with the concomitant loss of all that life holds and everything the victim holds dear. Experience of life and experience within the courts demonstrates that not all persons readily recover from such traumatic events. In other cases the fear may not be extreme and in not all cases will it continue to have an impact or significant impact on the victim after the threat causing it is removed. Not all people react to circumstances of stress in the same way. However, as I indicated above when referring to R v Broxam, R v Bell, and R v Stefanovski, the potential devastating psychological damage consequent on armed robberies is something of which the Courts have taken judicial notice over many years.
                325 By definition, the category of offence to which the guideline in this case is directed falls towards the bottom end of the scales of amount and premeditation. That is not to suggest these aspects are of no significance. The loss of $1,000 to a small shopkeeper may be more serious to him than the loss of $1M to a bank. The premeditation will rarely be of less than some hours duration and, even if the selection of the victim has not occurred long before the commission of an offence, the intention to commit, and preparation for, the offence will often be considerably longer than this.
                326 The category of offence does not fall so low when the other aspects to which I have referred are considered. The imposition on a shopkeeper is no different in kind from the imposition on a bank. The imposition on, fear engendered in, and the emotional or psychological reaction of the person threatened does not depend on whether they are a shopkeeper, bank teller or jewellery store owner. The risks of harm may be less with a knife than a gun, though this is not necessarily so and both can be fatal. Much will depend on often unpredictable actions of the victim and reactions of an offender who may be desperate for money and the drugs he can obtain with it. Indeed resistance by an uninsured shopkeeper faced with the loss of his own monies may be more likely than opposition from trained staff in larger organisations though one must recognise that in the latter situation there may be more persons exposed to danger.
                327 The commencing point in any consideration of an appropriate sentence for a statutory offence is the statute itself. As was said in another context, “In determining the proper penalty … the fundamental consideration is rather the degree by which, having regard to the maximum penalties provided by the Act in question, the Respondent’s conduct would offend against the legislative objective of suppressing the illicit traffic in the prohibited drug” - R v Peel (1971) 1 NSWLR 247 at 262.
                328 Having regard to the terms of the legislation to which I have referred and the nature of armed robbery, an offence which comes within a “worst case” category for which 20 years is the maximum penalty will commonly involve a large amount of money, a deal of premeditation, imposition on, and the risk of harm to, a number of people and serious psychological harm to at least one. It is likely it will involve “company” and/or an offensive weapon. The offender will commonly, though not necessarily, be a repeat offender. If one is considering a worst case for which 25 years is the maximum penalty, the offence is likely to involve also a dangerous weapon and the injury may be grievous bodily harm. Clearly, the category of case for which the guideline is intended falls a long way short of these sorts of offences. But so do the penalties I propose fall a long way short of 20 or 25 years. Although I do not suggest that anything shorter is insignificant, in absolute terms 5 years is but one quarter and 3 years is but 15% of the 20 year maximum. Terms of 3 and 5 years are shorter than the time spent by many people in tertiary or other training or in jobs which are of no long term significance to them. The periods are relatively short when compared with an adult life span. I said in R v Spiteri [1999] NSWCCA 3 that I regard 10 years imprisonment as more than twice as severe as a 5 years sentence The longer period is likely to have an impact on an offender’s life in terms of wife, children, job prospects and the like from which he may well never recover. Those observations apply a fortiori to a sentence of 20 or 25 years.
                329 In my view nothing less than the terms I proposed is appropriate as a norm or guideline for a premeditated offence of the nature of that described by the Chief Justice and which can be expected to place the victim, and possibly other members of the public, both in fear and at risk of serious injury or death and which is calculated, whatever the offender may intend, to cause significant psychological or other problems in at least part of the population. However small the amount taken, however short may have been the premeditation in a particular case, and however well an individual victim may have coped with the imposition on him or her, society has a great interest in protecting its members from the incidents necessarily attendant upon any armed robbery and in reducing the risks attendant on repetition of armed robberies, even small armed robberies, in the future. The punishment imposed must be such as to amount to a substantial deterrent.
                330 The statistics referred to when I was dealing with the topic of whether there should be a guideline judgment indicate, as I have said, that the offence of armed robbery has become far more common. The correct approach of the Court in that situation is set forth in the judgement of King CJ (with whom the other members of the court agreed) in R v Dube (1987) 46 SASR 118, quoted by the Chief Justice at paragraph 151 of his Reasons, viz. to increase the level of punishment. The increased incidence of the offence argues strongly that the guideline I propose is indeed insufficiently severe. However, as that guideline is a significant increase on the level of sentencing which has prevailed in the past, I am not persuaded that the guideline should be set at a level higher than I propose until one sees the result. After all, a significant number of offenders will receive a higher sentence than I have indicated because of the presence of aggravating circumstances such as a bad prior record and the commission of an offence while on conditional liberty.
                Drug Addiction
                331 I turn then to the topic of the significance of drug addiction in the determination of an appropriate penalty. On this topic I agree with the Reasons of the Chief Justice and Wood CJ at CL. In so doing, I should expressly acknowledge that I retreat from the remarks I made in R v Murray (unreported, CCA, 7 November 1995), quoted by Wood J. I am persuaded both by the reasoning of their Honours and by the authority quoted by them that, at least as a general proposition, what I said in R v Murray was wrong.

                IN THE COURT OF
                CRIMINAL APPEAL

                60559/98
                60558/98
                60561/98
                60746/98
                60596/98
                60595/98
                60511/98

                SPIGELMAN CJ
                WOOD CJ at CL
                NEWMAN J
                HULME J
                SIMPSON J

                                        Wednesday 12 May 1999

                REGINA v Paul Anthony HENRY
                REGINA v Stephen Anthony BARBER
                REGINA v Hoai Vinh TRAN
                REGINA v Troy David SILVER
                REGINA v Theo TSOUKATOS
                REGINA v Bill KYROGLOU
                REGINA v John David JENKINS

                GUIDELINE JUDGMENT


                SIMPSON J :
                332 I have read in draft the judgments of Spigelman CJ and Wood CJ at CL. I agree, for the reasons given by the Chief Justice, that it is appropriate that this court promulgate a guideline in relation to sentences imposed for offences against s 97 of the Crimes Act, and I agree with the guideline sentence proposed at para 165 of the draft judgment.
                333 I would, however, with respect to the offence/offender profile outlined in paragraph 162 of the draft judgment, express a preference for the promulgation of a guideline that relates to sentencing after conviction following a trial. (This would, of course, require lengthier terms than those specified in the proposed range.) There are three reasons for my preference. The first is that not all pleas of guilty carry the same weight, depending as they do on the strength of the Crown case, and the time at which they are offered. If the starting point assumed a plea of not guilty, then the appropriate discount to reflect the true worth of the plea could be given. The second reason is that the severity of the sentence to be imposed in relation to any offence might be mitigated by numerous facts or features, of which a plea of guilty is only one. I see no reason to single out the plea of guilty from other mitigating features in the definition of the kind of offence to which the guideline applies. The third and most substantial reason concerns appearances. Although the truth is that the guideline sentence has built into it a discount representing the benefit to which an offender is entitled for pleading guilty, the appearance, when an offender is sentenced following a trial, will be of a penalty that is attributed to the exercise of the right to the put the Crown to proof. The appearance would, as I have said, be false, but would exist nonetheless.
                334 However, as I understand that other members of the Bench do not share these concerns, I am content to concur in the guideline as proposed.
                335 I wish to say something about the submission advanced on behalf of certain of the respondents to the effect that drug addiction when a causal factor in the commission of offences should be recognised as a significant mitigating feature.
                336 It is a mistake, in my opinion, to regard drug addiction as a starting point. It is an end point, or a point on the way to the end, of a process. Where the process begins may vary in individual cases but it does not necessarily begin when a person decides to use a prohibited substance. It probably has its origins well before the date of the first use of drugs. In the worst, or least forgivable, cases it may have its origins in arrogance, in an antipathetical attitude to the laws of society, or in weakness of character. In other cases, I have no doubt, it has its origins in social disadvantage, poverty, emotional, financial, or social deprivation, poor educational achievement, unemployment, and the despair and loss of self-worth that can result from these circumstances or any combination of them. In this court one sometimes sees cases in which drug taking stems from sexual assault or exploitation, sometimes committed when the person who turns to drugs, and who comes before the court, is very young, and sometimes the precipitating events have occurred many years before. Drug addiction is not always the disease; it is, as often as not, a symptom of social disease.
                337 Drug addicts do not come to their addiction from a social or environmental vacuum. This court should not close its eyes to the multifarious circumstances of disadvantage and deprivation that frequently precede and precipitate a descent into illegal drug use. I do not suggest for a moment that all drug users fall into this category. It is because some do and some do not that I believe rigid rules about the impact on sentencing of drug dependency on sentencing cannot be laid down.
                338 I cannot accept that the blameworthiness of one drug taker is (even excepting that small number of individuals who begin drug taking with medically prescribed drugs) always to be treated as being at the same level as the blameworthiness of the next. Nor can I accept that the exercise of free choice in the use of drugs is always of equal dimensions. It is not every decision to use drugs that can properly or fairly be characterised as a decision made in the exercise of free choice. The will of an individual can be overborne or undermined, not only by acts of another person, but also by the pressure of circumstances. I do not accept that most drug offenders are truly exercising free will when they choose the degradation, despair, criminality and cycle of imprisonment that can follow the initial use of illegal drugs. The circumstances that propel the offender to the use of drugs are often, if not usually, beyond his or her control. They may or may not be combined with a vulnerable personality or even a weakness of character. Many drug offenders have not had the life experiences or the normal developmental path that permit a conclusion that the decision to take drugs was a decision made in the exercise of a free choice in the sense in which that phrase is ordinarily understood.
                339 An analogy can legitimately and properly be drawn with the principles relating to the sentencing of aboriginal offenders, collected and stated by Wood J (as he then was) in R v Fernando (1992) 76 A Crim R 58 at 62 - 63. As I understand the remarks on sentence in that case, his Honour observed that aboriginality alone does not excuse serious crime nor operate as a mitigating factor. However, since it is often associated with other circumstances of disadvantage or deprivation, aboriginality may explain or throw light on the particular offence or the circumstances of the offender. It is, in truth, not aboriginality, but the life experiences too commonly associated with aboriginality that are relevant to the sentencing decision. Just as it is properly said that, while drunkenness is not normally an excuse or mitigating factor, where alcohol abuse reflects the socioeconomic circumstances and the environment in which an offender has grown up, that fact can and should be taken into account as a mitigating factor (Fernando, principle E, p 62), so also can it properly be said that drug abuse may reflect the socioeconomic circumstances and the environment in which another offender has grown up, and that may equally then be taken into account as a mitigating factor.
                340 Wood J went on to recognise the great social difficulties sometimes faced by aboriginal communities where poor self image, absence of education and work opportunity and other demoralising factors have placed heavy stresses on them, reinforcing their resort to alcohol and compounding its worst effects. So, too, it is with some drug offenders. The passage from which the above is drawn can readily be adapted to the situation of some drug offenders. Where great social difficulties, poor self image, absence of education and work opportunity and other demoralising factors have placed heavy stresses upon drug offenders, causing or reinforcing their resort to drugs, then it is appropriate for this court to give the same recognition to those antecedent circumstances.
                341 A sentencing judge always has a delicate balancing task to perform, taking into account the demands of the community for retribution, deterrence both general and specific, and the interests of the same community in the rehabilitation of an offender. The exercise may call for an examination of the circumstances that led the offender to drug use, addiction and crime. All the circumstances that precipitate the use of drugs are relevant to the evaluation of moral culpability that is essential to the sentencing process.
                342 It would, therefore, be too simplistic to lay down a principle that addiction either is, or is not, a mitigating circumstance in the sentencing of offenders convicted of drug related crime. I much prefer the approach encapsulated in paragraphs 270 - 272 inclusive of the draft judgment of Wood CJ at CL, favouring the “individualised justice” to which Mahoney ACJ referred in R v Lattouf (unreported, NSWCCA, 12 December 1996).
                343 While, on this approach, it might be possible to argue that moral culpability for the offence is diminished, suggesting a more lenient rather than a more severe penalty, the counter balance is that, absent evidence of positive prospects of a cure of the drug addiction, a finding that the offence is unlikely to be repeated (itself ordinarily seen as a factor in favour of a lesser sentence) would be unavailable. Absent positive evidence of real rehabilitation, the diminution in moral culpability which results from drug addiction originating in the kind of social or familial deprivation to which I have referred cannot result in a corresponding diminution in sentence. Evidence of rehabilitation is the key to the impact evidence of drug addiction might have in the sentencing process. This view I perceive to be largely in accord with the direction taken by the legislature in the establishment of a Drug Court, specifically created to divert drug offenders with realistic prospects of rehabilitation (other than those involved in violence) from the criminal justice system: Drug Court Act 1998 (NSW) S 5(2)(b); para 4 of the draft judgment of Wood CJ at CL.
                344 Where circumstances such as those I have mentioned (or others equally deserving of compassion) have been the foundation for the drug addiction, and part of the causal chain leading to the commission of crime, then it would be appropriate, in my view, for the rehabilitative aspects of sentencing to assume a more significant role than might otherwise be the case. In an appropriate case, rehabilitation might outweigh other sentencing factors. In order for those circumstances to provide a reason for reduction of sentence, however, there would need to be strong evidence of real progress towards actual rehabilitation. I would not wish to be understood to be saying other than that leniency of the kind to which I refer depends heavily upon demonstrated (as distinct from theoretical) rehabilitative prospects.
                345 The authorities cited by Spigelman CJ and Wood CJ at CL are almost unanimous in holding that drug addiction of itself does not operate as a mitigating circumstance. When the words “of itself” are emphasised, I respectfully agree with the proposition repeatedly stated; but, in any event, the line of authority is so compelling and so sustained that, even sitting as one of a bench of five, I would be reluctant to participate in a decision departing from it. It is the role of the legislature (or the High Court) to alter a such a long standing principle. What I have said above is consistent with that line of authority . Nowhere is it held that it is inappropriate, in sentencing drug offenders, to take into account either the circumstances that gave rise to the drug addiction or demonstrated rehabilitation.
                346 In all the mass of material put before this court to support the argument that drug addiction should be treated as a mitigating factor, there was nothing to inform the court, statistically or otherwise, of what causes drug addiction, or its socioeconomic background. What I have said, therefore, is somewhat impressionistic, much of it drawn from experience on this court, and some of it assumptions based upon more general reading. It is, I believe, largely uncontroversial. It is not intended as a recitation of the most universal or common circumstances pre-dating or causing drug addiction, but of some familiar patterns.
                347 I have not lost sight of the material to which the court was taken concerning the effect of armed robberies on their victims. Obviously, the offence is serious, commonly has drastic consequences for victims, and must be punished accordingly.
                348 I agree that the principles stated by Wood CJ at CL in paragraph 273 of his draft judgment are the principles that emerge from the authorities. To them, I would add that not all drug addicts are necessarily to be taken to have brought themselves to addiction and criminality with the same degree of foresight, determination, informed decision making, and consequent moral culpability. To impute those qualities to every decision to take drugs is to ignore the important starting point, the reasons underlying the initial drug use.
                349 I entirely agree with the proposition that the bare fact that an offence is motivated by a need for money to support a drug habit does not, alone, mitigate the offence or operate to reduce the sentence to be imposed.
                350 Counsel who appeared for some respondents also argued, by analogy with the sentencing of mentally abnormal offenders, that, in sentencing drug addicted offenders, it may be appropriate to afford less weight to the principle of general deterrence. This argument cannot be accepted. In considering general deterrence it is necessary to consider the class or pool of individuals at whom deterrence is directed. In s 97 offences, that class or pool is the class of people who commit, or are tempted to commit, the offence of armed robbery - many of whom, as the materials show, are also motivated by the need for money to feed drug addiction. These are people whose will not to commit crimes is already weakened by the drug dependence. That means an even stronger message must be conveyed. Far from diminishing the importance of general deterrence, the fact that the population in which deterrence is necessary is unlikely to be deterred other than by firm action on the part of the courts speaks for greater rather than lesser emphasis on that principle.
                351 Usually, and certainly in relation to armed robbery, general deterrence is a very significant factor in the sentencing process. So is specific deterrence and so is punishment. The point I wish to make is that where a combination of two circumstances exists, then general deterrence, and other sentencing objectives such as retribution, may yield to rehabilitation. The two circumstances are:
                352 (i) a background to the addiction that may explain the offender’s decision to use drugs, and diminish his or her moral culpability for that decision;
                353 (ii) demonstrated progress towards rehabilitation.

                354 The second is, to my mind, essential before general deterrence can be seen to give way to rehabilitation.
                355 Where those two circumstances coincide, then the interests of the community may well be better served by the imposition of a penalty that leans towards furthering the rehabilitative process at the expense of the punitive and even the deterrent objectives of sentencing.
                356 I do not understand this approach to be significantly different from the recognition given by Wood CJ at CL to the need for sentencing to be sufficiently flexible to take account of all relevant circumstances (paragraphs 270 - 2719). It is consistent with many authorities in this court and other jurisdictions: see, eg R v Lattouf (unreported, NSW CCA, 12 December 1996); R v Hayes (1987) 29 A Crim R 452 at 457, 472; R v Eager (unreported, NSW CCA 1 November 1995); R v Crotty (unreported, NSW CCA 28 February 1994); R v Fabian (1992) 64 A Crim R 365 at 372, 378, 380; R v Molina (1984) 13 A Crim R 76 at 77 (Federal Court of Australia) (quoted with approval in R v Ellis (1993) 68 A Crim R 449 by Kirby P, who, however, dissented in the result in that case); see also the remarks in Molina in relation to the offender there before the court at p 79.2; R v Dowie (1989) 42 A Crim R 234 at 247 (Court of Criminal Appeal, Tasmania); R v Osenkowski (1982) 30 SASR 212 at 212-3 (Supreme Court of South Australia in Banco); R v Halewyn (1984) 12 A Crim R 202 at 205-6 (Court of Criminal Appeal, Victoria).
                **********
                - 3 -
                IN THE COURT OF
                CRIMINAL APPEAL


                60559/98
                60558/98
                60561/98
                60746/98
                60596/98
                60595/98
                60511/98


                                            SPIGELMAN CJ
                                            WOOD CJ at CL
                                            NEWMAN J
                                            HULME J
                                            SIMPSON J


                                            Wednesday 12 May 1999

                            REGINA v Paul Anthony HENRY
                            REGINA v Stephen Anthony BARBER
                            REGINA v Hoai Vinh TRAN
                            REGINA v Troy David SILVER
                            REGINA v Theo TSOUKATOS
                            REGINA v Bill KYROGLOU
                            REGINA v John David JENKINS

                Six Crown appeals and one severity appeal involving the offence of armed robbery under s97(1) (one involving the aggravated offence under s97(2)) of the Crimes Act 1900 (NSW) were heard together. The Crown submitted that it was appropriate to promulgate a guideline judgment with respect to this offence.

                Held (By the Court):
                Sentencing Guidelines: General
                The purpose of a guideline judgment is to foster consistency in sentencing: Jurisic NSWCCA 12 October 1998; Lowe (1994) 154 CLR 606 at 610-611.

                The appropriateness of an appellate court promulgating guidelines has been authoritatively established: Norbis v Norbis (1986) 161 CLR 513.

                A guideline judgment on the subject of sentencing does not lay down anything in the nature of a rule. The failure to sentence in accordance with a guideline is not itself a ground of appeal: Norbis per Brennan J at 536, 537-538; Jurisic at pp15,16.

                Consideration of the Canadian and South Australian approaches to sentencing guideline. McDonnell (1997) 114 CCC (3d) 4336; Police v Cadd (1997) 94 ACrimR 466; Bini (1994) 68 ALJR 859, referred to.

                Use of New Material in Individual Cases
                Consideration of the use of material before the appellate court that was not before the sentencing judge. The word “evidence” in s12 of the Criminal Appeal Act 1912 and s29A of the Criminal Procedure Act 1986 is confined to “matters going to the commission of the offence or the personal circumstances of the accused”: Beldan (1986) 21 ACrimR 159; Chanh Nghia Ly NSWCCA 16 December 1992; J (1992) 64 ACrimR 441 at 459; Veen v The Queen (No 2) (1987-88) 164 CLR 465; Hallocoglu (1991) 29 NSWLR 67.

                Therefore criminal statistics on the incidence of an offence, including police data, court data and victim surveys are not “evidence” within s12 of the Criminal Appeal Act 1912 nor s29A of the Criminal Appeal Act 1986. Whilst academic literature on the effects of an offence including both physical and psychological effects, may not be “evidence” within these sections, (Hulme J dissenting), it is not material to which the Court should have regard in the individual cases as the gravity of each case is determined by, inter alia, the effects on the particular victims.

                Per Hulme J:
                Observations on criminal statistics.

                Need for Guidelines: Armed Robbery
                The increased incidence of a particular offence is a relevant factor when considering whether or not a guideline judgment is appropriate and, if so, what change with respect to past sentencing practice is called for. Whilst armed robbery is a relatively rare crime, its rate of occurrence has significantly increased in recent years.

                Furthermore, the sentencing statistics of the Judicial Commission of New South Wales indicate:
                (i) Non-custodial sentences are not confined to exceptional cases
                (ii) Leniency is suggested in the full terms
                (iii) The pattern of leniency is reinforced by the extent to which the statutory ratio between minimum and additional terms has been varied
                ((iv) The large proportion of lenient sentences suggest inconsistency in sentencing practices.

                The sentencing statistics suggest both inconsistency and systematic leniency, which justify the promulgation of a guideline judgment.

                Sentencing Guidelines: Armed Robbery
                (a) Type of penalty
                Armed robbery is not simply a crime against property. It is a crime against persons. The fear engendered by the perpetrator of this crime, together with the continued adverse effects on its victims, establish armed robbery to be a serious crime which required condign punishment. Non-custodial sentences for this offence should only be imposed in exceptional circumstances: Roberts (1994) 73 ACrimR 306 at 308. The number of cases in which non-custodial sentences have been imposed indicates that this principle has not been implemented by sentencing judges. The sentencing statistics suggest that this approach is also reflected in the length of custodial sentences when imposed. (Waldron NSWCCA 3 March 1994. See also Ellis (1993) 68 ACrimR 449 at 462; Lawson NSWCCA 12 December 1997).

                (b) Guideline precedents
                Guidelines for armed robbery sentencing in English, New Zealand, Hong Kong, Canadian, Western Australian, South Australian and Queensland authorities referred to. (Turner (1975) 61 CrAppR 67; Gould (1983) 5 CrAppR(S) 72; Attorney-General’s References Nos 3, 4, 8, 9, 10, 11 and 16 of 1990 (Dickson & Ors) (1991) 92 CrAppR 166; Attorney-General’s Reference No 14 of 1991 (Reed) (1992) 13 CrAppR(S) 446; Attorney-General’s Reference NO 7 of 1992 (Khan) (1993) CrAppR(S) 122; Moananui (1983) NZLR 537; Mo Kwong Sang (1981) HKLR 610; Brennan and Jensen (1975) 11 NSR (2d) 541; Chaisson (1975) 24 CC (2d) 159; Johnas (1982) 2 CCC (3D) 490; Norman WACCA 1 February 1989; Miles (1997) 17 WAR 518; Spiero (1979) 22 SASR 543; Dube (1987) 456 SASR 118; Fermaner (994) 61 SASR 447; Drumgoon SACCA 20 November 1995; Hammond (1996) 92 ACrimR 450).

                (c) Guidelines
                A category of case which is sufficiently common for the purposes of determining a guideline comprises the following elements:
                (i) Young offender with no or little criminal history
                (ii) Weapon like a knife, capable of killing or inflicting serious injury
                (iii) Limited degree of planning
                (iv) Limited, if any, actual violence but a real threat thereof
                (v) Victim in a vulnerable position such as a shopkeeper or taxi driver
                (vi) Small amount taken
                (vii) Plea of guilty, the significance of which is limited by a strong Crown case.

                Per Simpson J:
                A plea of guilty should not be included in the definition of the kind of offence to which the guideline applies.

                (d) The appropriate range
                A sentencing range is appropriate in relation to this offence because the seven identified characteristics do not exhaust the factors relevant to sentencing, and many of the characteristics contain within themselves an inherent variability.

                Sentences for an offence of this character should generally fall between four and five years for the full term. Aggravating and mitigating factors will justify a sentence below or above the range, which is itself a starting point.

                Per Hulme J:
                The guideline should be set so as to include a total term of penal servitude of 5 years and, except in unusual circumstances a minimum term of 3 years.

                (e) Circumstantial factors
                A number of circumstances are particular to the offence of armed robbery, including: (a) nature of the weapon; (b) vulnerability of the victim; (c) position on a scale of impulsiveness/planning; (d) intensity of threat, or actual use, of force; (e) number of offenders; (f) amount taken; and (g) effect on victim/s.

                Per Spigelman CJ:
                Drug addiction is a circumstance relevant to the sentencing exercise, but it is not itself a mitigating factor. The existence of a causal relationship between drug addiction an the commission of an offence should not automatically result in a lesser sentence. Valentini (1989) 46 ACrimR 23 at 25; Halewyn (1984) 12 ACrimR 202; Nolan [1998][ VSCA 135; Douglas (1995) 56 FCR 465 at 470; Spiero (1979) 22 SASR 543 at 549; Terizakis (1986) 41 SASR 252 at 256; Hammond (1996) 92 ACrimR 450 at 455-456, 467.

                Per Wood CJ at CL:
                Discussion of general principles of deterrence and drug addiction. Drug addiction is not analogous to mental abnormality in respect of which the element of general deterrence may be given less weight. The relevant principles are as follows:
                (a) The need to acquire funds to support a drug habit is not an excuse to commit an armed robbery, and of itself not a matter of mitigation.
                (b) The fact that an offence is motivated by such a need may be taken into account as a factor relevant to the objective criminality of the offence, insofar as it may throw light on matters such as (i) the impulsivity of the offence and the extent of any planning for it; (ii) the existence or non-existence of any alternative reason that may have operated in aggravation of the offence; and (iii) the state of mind or capacity of the offender to exercise judgment.
                (c) It may also be relevant as a subjective circumstance, insofar as the origin or extent of the addiction, and any attempts to overcome it, might (i) impact upon the prospects os recidivism/rehabilitation; (ii) suggest that the addiction was not a matter of personal choice but was attributable to some other event for which the offender was not primarily responsible; and (iii) justify special consideration in the case of offenders to be at the “cross roads”: Osenkowski (1982) 5 ACrimR 394.

                Per Simpson J:
                Discussion of general principles of rehabilitation and drug addiction. Where a combination of the following two circumstances exist, then general deterrence and other sentencing objectives such as retribution, may yield to rehabilitation: (i) a background to the addiction that may explain the offender’s decision to use drugs, and diminish his or her moral culpability for that decision; (ii) demonstrated progress towards rehabilitation.
                - 139 -
                IN THE COURT OF
                CRIMINAL APPEAL


                60559/98
                60558/98
                60561/98
                60746/98
                60596/98
                60595/98
                60511/98


                                            SPIGELMAN CJ
                                            WOOD CJ at CL
                                            NEWMAN J
                                            HULME J
                                            SIMPSON J


                                            Wednesday 12 May 1999

                            REGINA v Paul Anthony HENRY
                            REGINA v Stephen Anthony BARBER
                            REGINA v Hoai Vinh TRAN
                            REGINA v Troy David SILVER
                            REGINA v Theo TSOUKATOS
                            REGINA v Bill KYROGLOU
                            REGINA v John David JENKINS

                JUDGMENT

                1 SPIGELMAN CJ: In Jurisic NSWCCA 12 October 1998, this Court indicated that it would in future be prepared to issue guideline judgments with respect to sentencing for particular offences. The Court has listed and heard together six Crown appeals and one severity appeal, six of which involves the offence of armed robbery under s97(1) of the Crimes Act 1900 (NSW) and one, the aggravated offence under s97(2). The Crown has submitted that it is appropriate to promulgate a guideline judgment with respect to this offence.
                2 In Jurisic, in a judgment with which the other four members of the Court agreed, I said:
                “In my opinion, guideline judgments should now be recognised in New South Wales as having a useful role to play in ensuring that an appropriate balance exists between the broad discretion that must be retained to ensure that justice is done in each individual case, on the one hand, and the desirability of consistency in sentencing and the maintenance of public confidence in sentences actually imposed, and in the judiciary as a whole, on the other.” (14)
                    “Such guidelines are intended to be indicative only. They are not intended to be applied to every case as if they were rules binding on sentencing judges. Decisions of appellate courts on sentencing are not to be treated as binding precedents.” (14-15)
                3 In Jurisic I referred to the numerous cases in which this Court has made statements of general principle with a view to guiding sentencing practice in trial courts. The promulgation of formal guideline judgments, labelled as such, was a development of that practice.
                4 In Jurisic, Wood CJ at CL said:
                “The Court has … over the years endeavoured to lay down sentencing principles for particular classes of case where sentences reflecting a significant element of general deterrence are required, or when non-custodial options are inappropriate. It appears that sometimes these principles are lost or that their significance is overlooked, in the volume of appellant decisions handed down and in the pressures imposed on trial courts to dispose of increasingly busy criminal lists.
                    By tagging selected decisions as guideline judgments, the Court is not to be taken as usurping the function of the legislature, or as inappropriately intruding into the exercise of the sentencing discretion reserved to trial judges. Rather, what is intended is for the Court of Criminal Appeal to highlight the sentencing principles which fall for it to determine, in a way that might assist trial judges, the DPP and trial counsel, and reduce the occasion for that degree of inconsistency or departure from principle that is an indicator of injustice.” (1-2)
                5 Subject to any relevant statutory requirements, the sentencing task involves the exercise of a broad discretion which, centuries of practical experience strongly indicate, is best conferred on trial judges.
                6 The circumstances in which it is appropriate for appellate courts to interfere with discretionary decisions of this character are confined. In the case of Crown appeals against sentence, even more stringent restrictions have been applied.
                7 The ineluctable core of the sentencing task is the process of balancing overlapping and contradictory objectives. At the appellate level, that characteristic extends to the balancing of the objectives of consistency and individualisation. The Court must sentence both the offender and the offence.
                8 During the course of these proceedings the Respondents to the Crown appeals and the Appellant in the severity appeal, relied on certain observations of Mahoney ACJ in Lattouf (NSWCCA 12 December 1996) where his Honour repeated his own remarks in Kable v DPP (1995) 36 NSWLR 374 at 394:
                “If justice is not individual, it is nothing.”
                9 This ringing phrase must not be taken out of context. In Lattouf his Honour emphasised the multiple objectives served by the sentencing process. One could equally well say “If justice is not consistent, it is nothing”.
                10 As His Honour put it in Lattouf:
                “General sentencing principles must be established, so that the community may know the sentences which will be imposed and so that sentencing judges will know the kind and the order of sentence which it is appropriate that they impose. But, of course, principles are necessarily framed in general terms. General principles must, of their nature, be adjusted to the individual case if justice is to be achieved. For this reason, it is in my opinion important in the public interest that the sentencing process recognise and maintain a residual discretion in the sentencing judge … There is a public interest in the adoption and articulation of sentencing principles which will deter the commission of serious crime and punish those who commit it … But there are other interests to which the sentencing process must have regard; these are other objectives which the sentencing process must seek to achieve. Paramount amongst these is the achievement of justice in the individual case.” (7)
                11 There is no conflict between the system of guideline judgments established by Jurisic and the reasoning of Mahoney ACJ in Lattouf.


                Guidelines: General
                12 As I indicated in Jurisic, the purpose of a guideline judgment is to foster consistency in sentencing. The importance of consistency was well expressed by Mason J in Lowe (1994) 154 CLR 606:
                “Just as consistency in punishment - a reflection of the notion of equal justice - is a fundamental element in any rational and fair system of criminal justice, so inconsistency in punishment, because it is regarded as a badge of unfairness and unequal treatment under the law, is calculated to lead to an erosion of public confidence in the integrity of the administration of justice.” (610-611)
                13 The appropriateness of an appellate court establishing guidelines has been authoritatively established. In Norbis v Norbis (1986) 161 CLR 513, the High Court had before it an issue concerning the power of the Full Court of the Family Court to lay down guidelines with respect to the exercise of statutory discretions by trial judges. By majority, the Court held that the promulgation of such guidelines was permissible.
                14 Mason and Deane JJ said:
                “It has sometimes been said by judges of high authority that a broad discretion left largely unfettered by Parliament cannot be fettered by the judicial enunciation of guidance in the form of binding rules governing the manner in which the discretion is to be exercised. However, it does not follow that, because a discretion is expressed in general terms, Parliament intended that the Court should refrain from developing rules or guidelines affecting its exercise. One very significant strand in the development of the law has been the judicial transformation of discretionary remedies into remedies which are granted or refused according to well settled principle. It has been a development which has promoted consistency in decision making and diminished the risks of arbitrary and capricious adjudication. The proposition referred to in the beginning of this paragraph should not be seen as inhibiting an appellate court from giving guidance, which falls short of constituting a binding rule, as to the manner in which the discretion should be exercised. And despite the generality of some of the statements to which we have referred, there may well be situations in which an appellate court will be justified in giving such guidance the force of a binding rule by treating a failure to observe it as constituting grounds for a finding that the discretion has miscarried.
                    The point of preserving the width of the discretion which Parliament has created is that it maximises the possibility of doing justice in every case. But the need for consistency in judicial adjudication, which is the antithesis of arbitrary and capricious decision making, provides an important countervailing consideration supporting the giving of guidance by appellate courts, whether in the form of principles or guidelines. The tension between the two considerations, each of fundamental importance in family law, has inevitably led to a near dilemma for the Full Court of the Family Court. To avoid the risk of inconsistency and arbitrariness which is inherent in the system of relief involving a complex of discretionary assessments and judgments, the Full Court, as a specialist appellate court with unique experience in family law in this country, should give guidance as to the manner in which these assessments and judgments are to be made. Yet guidance must be given in a way that preserves, so far as it is possible to do so, the capacity of the Family Court to do justice according to the needs of the individual case, whatever its complications may be. Reconciliation of these goals suggests that in most, if not all, cases the Full Court of the Family Court should give guidance in the form of guidelines rather than binding principles of law.” (519-520 internal references omitted)
                15 The other member of the majority in Norbis was Brennan J. He agreed with the reasons of Mason and Deane JJ with one exception:
                “The proposition with which I am unable to agree is this: that an appellate court that gives guidance as to the manner in which a statutory discretion should be exercised may prescribe that such guidance should have the force of a binding legal rule. The authority of an appellate court to give guidance is not to be doubted. It is inevitable that the wisdom gained in continually supervising the exercise of a statutory discretion will find expression in judicial guidelines. That is not to invest an appellate court with legislative power but rather to acknowledge that, in the way of the common law, a principle which can be seen to be common to a particular class of case will ultimately find judicial expression. The orderly administration of justice requires that decisions should be consistent one with another and decision making should not be open to the reproach that it is adventitious … An unfettered discretion is a versatile means of doing justice in particular cases, but unevenness in its exercise diminishes confidence in the legal process.” (536)
                16 His Honour went on to say:
                “It is one thing to say that principles may be expressed to guide the exercise of a discretion; it is another thing to say that the principles may harden into legal rules which would confine the discretion more narrowly than the Parliament intended.” (537)
                17 His Honour also said:
                “There may well be situations in which an appellate court will be justified in setting aside a discretionary order if the primary judge, without sufficient grounds, has failed to apply a guideline in the particular case. Where there is nothing to mark the instant case as different from the generality of cases, the failure will suggest that the discretion has not been soundly exercised. The distinction between such a guideline and a binding rule of law, though essential, may be thin in practice. But the distinction must be maintained and a failure to apply the guideline cannot be treated as an error of law: a failure to apply the guideline is no more than a factor which warrants a close scrutiny to the particular exercise of the discretion. What cannot be shut out is the discretion of a primary judge not to apply the guideline when the circumstances of the particular case show that its application would produce an unjust or inequitable result or that another approach would produce a more just and equitable result.
                    The only compromise between idiosyncrasy in the exercise of the discretion and an impermissible limitation of the scope of the discretion is to be found in the development of guidelines from which a judge may depart when it is just and equitable to do so - guidelines which are not rules of universal application, but which are generally productive of just and equitable orders. If it is possible to develop such guidelines, it is possible to ensure order and consistency in the exercise of the discretionary jurisdiction under the Family Law Act.” (537-538)
                18 The appropriateness of an appellate court laying down guidelines for the exercise of a statutory discretion has also been noted in a range of cases, invariably with reference to the desirability of consistency in judicial decision making. These cases include:
                · the statutory discretion to award costs in criminal proceedings (Latoudis v Casey (1990) 170 CLR 534 at 541-542, 558-559, 562).
                · the exercise of the discretion to award costs in civil matters (Oshlack v Richmond River Council [1998] HCA 11; 72 ALJR 578; 152 ALR 83 at [35], [134] and [65])
                · the exercise of a discretion to order the winding up of a company (FAI Insurances Limited v Goldleaf Interior Decorators Pty Ltd (No 2) (1988) 14 NSWLR 644 at 646-647, 657-658 and 660-661).
                · the exercise by a licensing magistrate of a discretion to grant a liquor licence (Shreeve v Martin (1969) 72 SR(NSW) 279 at 289-291).
                · the exercise of a statutory discretion to grant leave to commence proceedings out of time (Salido v Nominal Defendant (1993) 32 NSWLR 524 at 532-533, 535-539, 541).
                · The exercise of a discretion to grant leave to appeal from decisions of arbitrators (Pioneer Shipping Limited v BTP Tioxide Limited [1982] AC 724 at 742-743; FAI Insurances v Goldleaf Interior Decorators supra at 661; Leighton Contractors Pty Limited v Kilpatrick Green Pty Limited [1992] 2 VR 505 at 516, 517, 521).
                19 The High Court has confirmed that a Court may establish guidelines for the exercise of an inherent jurisdiction, namely the parens patriae jurisdiction to act in the best interest of the child. (Secretary, Department of Health and Community Services v JWB and SMB (1991) 175 CLR 218 at 260).
                20 The Full Court of the Supreme Court of Victoria has applied the approach to guidelines in Norbis to the determination of guiding principles for the exercise of a discretionary power of the Court in its inherent jurisdiction, namely the power to order the dismissal of proceedings for want of prosecution. (Maysell v Transport Industries Insurance Co Limited [1995] 2 VR 328 at 334-335).
                21 In the case of matters covered by common law principles - like sentencing for New South Wales offences - there is no inhibition which may arise from the prospect of confining a statutory discretion, which the Parliament has conferred in an unconfined form. (Special considerations will arise in formulating guidelines for Commonwealth offences by reason of s16A of the Crimes Act 1914).
                22 As noted above, in Norbis the joint judgment of Mason and Deane JJ contemplated the possibility of a guideline having “the force of a binding rule”, with the consequence that failure by a trial judge to apply a guideline may constitute a ground for finding that the exercise of the discretion miscarried. It was in this respect that Brennan J disagreed.
                23 The Full Court of the Family Court has adopted the option contemplated by Mason and Deane JJ:
                “In our view the time has come to regard a departure from a longstanding guideline … without adequate explanation as a ground for finding that the exercise of discretion has miscarried.” Docters Van Leeuwen (1990) 14 FamLR 130 at 134.
                    (See also Lalor (1989) 14 FamLR 282 at 285; Joshua (1997) 22 FamLR 203 at 214; Nygh “Should the Full Court Offer More Guidance to Judges Sitting at First Instance” (1993) 7 Australian Journal of Family Law 137 esp at 146-149).
                24 However, a differently constituted Full Court of the Family Court has stated that guidelines developed by the Full Court should not be regarded as rules of law:
                “The guidelines that we propose are simply guidelines; they are not rigid rules of law and it does not follow that a departure from them will necessarily vitiate a judgment, although judges, judicial registrars and registrars should, we think, give sufficient reasons for departing when they consider a departure is appropriate.” (Re K (1993) 117 FLR 63 at 81).
                25 The balance of authority strongly supports the reasoning of Brennan J in Norbis. See Shreeve v Martin supra at 290 per Walsh JA; FAI Insurances v Goldleaf Interiors supra at 661 per McHugh JA; Latoudis v Casey supra at 558-559 per Dawson J, with whom Brennan J agreed; Leighton Contractors v Kilpatrick Green supra at 516-517 per Fullagar J with whom McGarvie and Ashbee JJ agreed; Maysell v Transport Industries Insurance supra at 334-335 per Brooking, Teague and Hedigan JJ; Oshlack v Richmond River Council supra at [35] per Gaudron and Gummow JJ, and at [134] per Kirby J.
                26 A similar issue has arisen in Canada. The Supreme Court of Canada has unanimously affirmed that it is appropriate for a criminal appellate court to lay down guidelines in the nature of a starting point for sentencing for a particular offence. McDonnell (1997) 114 CCC (3d) 436 at pars 43, 58-61. (See generally Bloos and Renke “Case Comment: Stopping Starting Points R v McDonnell” (1997) 35 Alberta LR 795).
                27 In McDonnell both (majority and minority) judgments rejected the proposition that departure from a previously decided starting point could, itself, constitute an error of principle that would justify appellate intervention ( at pars 32-34; 100, 109).
                28 The majority accepted that departure from a starting point may “suggest” that a sentence is outside the permissible range but is not “determinative” of that proposition (par 43). The minority seemed to regard departure from the starting point as being more strongly indicative of legal error, in terms of manifest inadequacy (or “demonstrably unfit” in the Canadian terminology) (pars 98-101). Neither judgment adopted the approach which Mason and Deane JJ contemplate in Norbis.
                29 A guideline judgment on the subject of sentencing should not lay down a requirement or anything in the nature of a rule. The failure to sentence in accordance with a guideline is not itself a ground of appeal. Guidelines are not rules of universal application. They may be departed from when the justice of a particular case requires such departure. I made this clear in Jurisic when I said:
                “Such guidelines are intended to be indicative only. They are not intended to be applied in every case as if they were rules binding on sentencing judges. Decisions of appellate courts on sentencing are not to be treated as binding precedents.” (15)
                30 In this respect I refer to my adoption in Jurisic (at 15) of the analysis by Dunn LJ in De Havilland (1993) 5 CrAppR(S) 109 at 114, to the effect that decisions on sentencing are not authorities binding on lower courts in the way decisions on substantive law are binding. I went on to say:
                “ … such guidelines are not binding in any formal sense. They represent a relevant indicator, much as trial judges have always regarded statutory maximum penalties as an indicator”. (Jurisic at 16)
                31 Nevertheless, where a guideline is not to be applied by a trial judge, this Court would expect that the reasons for that decision be articulated, so that the public interest in the perception of consistency in sentencing decisions can be served and this Court can be properly informed in the exercise of its appellate jurisdiction.
                Guidelines: Sentencing
                32 In Jurisic I outlined the development of sentencing guideline judgments in England. It is not necessary to repeat that analysis.
                33 In Canada, a starting point approach to sentencing guidelines has been endorsed by the Supreme Court in McDonnell supra. McLachlin J said:
                “The starting point approach to sentencing involves two steps. First, the judge determines the range of sentences for a typical case. Using that range as a starting point, a trial judge then adjusts the sentence upward or downward on the basis of factors relating to the particular offence and offender … the starting point approach combines general considerations relating to the crime committed with personalised considerations relating to the particular offender and the unique circumstances of the assault.” (par 58)
                34 Her Ladyship went on to say:
                “The traditional notion that sentencing is primarily a matter of impression for the sentencing judge and only secondarily a matter of principle began to be questioned by the Courts in the mid 60s. Behind the challenge lay increasing recognition that some measure of uniformity was essential in a sentencing process that was not only just, but was perceived to be just.” (par 65)
                    and
                “The starting point approach appears to meet both the requirements of uniformity and consistency in sentencing and individualised justice.” (par 78)
                    (See also pars [23-24, 43]; Ruby Sentencing (4th ed, 1994) esp pp481-482).
                35 The judgments in the Supreme Court referred with approval to the decision of the Alberta Court of Appeal in Sandercock (1985) 22 CCC (3d) 79 where the Court said:
                “The sentencing process now adopted by this court is to state typical categories with precision, and to acknowledge at the same time that each actual case presents differences from the archetypical case. These differences might mitigate or aggravate.” (83)
                36 This is similar to the approach adopted by this Court in Jurisic.
                37 The practice of the Full Court of the Supreme Court of South Australia is to promulgate sentencing standards in the form of an “appropriate sentence range”. However, departure from the range is not itself error.
                38 In Police v Cadd (1997) 94 ACrimR 466, Doyle CJ said:
                “It is the function of this Court to ensure that the sentencing process in individual cases is not affected by error of principle. It is also the function of this Court to ensure that sentences are neither excessive nor inadequate. The latter function is performed in two ways. First, in individual cases, by correcting a particular sentence that is considered to be excessive or inadequate. Secondly, by establishing standards of sentencing for particular offences, when the court thinks it appropriate to do so. This may be done over time through the process of correcting individual sentences. But it may also, in my opinion, be done by the court indicating an appropriate sentence range for a particular offence or offences of a particular type.” (479-480)
                39 His Honour went on to cite a number of cases in which the Court had indicated appropriate sentencing ranges and referred, with approval to the following remarks of Cox J in King (1988) 48 SASR 555 at 557-558; 34 ACrimR 412 at 414-415:
                “… this case is about sentencing standards, but it is important, I think, to bear in mind that when a sentence is created, either by cumulative force of individual sentences or by a deliberate act of policy on the part of the Full Court, there is nothing rigid about it. Such standards are general guides to those who have to sentence in the future …”
                40 Doyle CJ went on to say in Cadd:
                “The function of establishing appropriate standards is an important aspect of ensuring, as best one can, that adequate standards of punishment are observed. Establishing appropriate standards also tends to ensure that there is such consistency of approach as is achievable in a system in which the appropriate sentence depends upon, in part, the circumstances of the individual case and of the individual offender, and a system in which sentencing is as individualised as it is in our system.” (480)
                    (See also 487 per Duggan J; 488 and 490-491 per Mullighan J; 511 per Lander J; 520 per Bleby J).
                41 I express my agreement with the following observations of Lander J:
                “There are a number of reasons why this Court can and should indicate in a general way that a particular class of offences should be dealt with within certain standards. First it gives general guidance to judges of this Court and to the courts below of what the superior court within the State considers an appropriate range of penalties for a particular class of offence. Secondly an indication by this Court of an appropriate standards is likely to best ensure consistency of penalty throughout the State. Thirdly it is, in my opinion, part of the deterrent process that this Court pronounces, not only for the assistance of the courts below, but for the general education of the public and the particular education of those who may be likely to offend. The public should know in advance that offences of a particular kind will be likely to lead to a penalty in a range indicated by this Court.
                    I agree therefore that it is appropriate for this Court to indicate a standard in relation to some offences. However not all offences are capable of being categorised in that manner. Some offences such as assaults and manslaughter depend so very much upon the circumstances of the particular offence that they cannot be considered in a general way as being within any particular sentence range.” (511)
                42 The South Australian practice was affirmed by the High Court when rejecting a special leave application from one of four sentencing standard cases which had been heard together and which I will consider further below. Brennan J said:
                “By their decisions, the Court of Criminal Appeal hoped to ‘remind sentencing judges of the great importance of maintaining adequate standards of punishment in sentencing for armed robbery’. That being the object of the Court of Criminal Appeal, this Court should not grant special leave to review the range of sentences which the Court of Criminal Appeal has set.
                    However, the Court of Criminal Appeal is bound to apply general principles of sentencing to any case in which the Crown seeks to have a range of sentencing established or confirmed so that the actual sentence in any case properly reflects its unique circumstances.” (Bini (1994) 68 ALJR 859)
                    Statutory Scheme for Robbery
                43 Chapter 1 of Pt 4 of the Crimes Act 1900 (NSW) creates a hierarchy of offences with respect to robbery, an offence distinguished from other forms of stealing by an element of threat or force, putting a person in fear:
                Maximum Penalty 14 years
                    (i) Robbery or assault with intent to rob from the person of another (s94).
                    Maximum Penalty 20 years
                    (ii) The offence in (i), in circumstances of aggravation, which includes use of violence or malicious infliction of actual bodily harm or deprivation of liberty (s95).
                    (iii) Robbery or assault with intent to rob being armed with an offensive weapon or in company (s97(1)).
                    Maximum Penalty 25 years
                    (iv) The offence in (ii), with actual wounding or infliction of bodily harm (s96).
                    (v) The offence in (iii), when armed with a dangerous weapon (s97(2)).
                    (vi) Wounding or inflicting grievous bodily harm before, at the time of or immediately after a robbery, when armed with an offensive weapon (s98).
                44 In Brown (1989) 17 NSWLR 472 in a joint judgment of Gleeson CJ, Newman and Loveday JJ, this Court said:
                “As is the case in relation to various other subject matters, the provisions of the Crimes Act 1900 dealing with robbery establishes a series of offences, in ascending degrees of seriousness, and with ascending orders of maximum penalty, depending upon the circumstances of the case … It is apparent that within that range of offences there may exist an enormous variety of individual facts and circumstances which will accompany the commission of a particular crime.” (473)
                45 The cases now before the Court are concerned only with s97(1), save for one conviction under s97(2). The section provides:
                “97(1) Whosoever, being armed with an offensive weapon, or instrument, or being in company with another person, robs or assaults with intent to rob, any person, or stops any mail, or vehicle, railway train, or person conveying a mail, with intent to rob, or search the same, shall be liable to penal servitude for 20 years.
                    (2) A person is guilty of an offence under this subsection if the person commits an offence under subsection (1) when armed with a dangerous weapon. A person convicted of an offence under this subsection is liable to penal servitude for 25 years.”
                46 Section 4(1) defines “offensive weapon or instrument” to:
                “… include a dangerous weapon and also include an imitation or replica of an offensive weapon or of an offensive weapon, or an instrument as the case may require.”
                47 “Dangerous weapon” means a firearm, a prohibited weapon within the Prohibited Weapons Act 1989 (NSW) or a spear gun.
                48 It is with respect to the offence under s97(1) that the Court was invited by the Crown to promulgate a guideline judgment.
                Crown Material
                49 The Crown sought to put before the Court a range of materials which were not before the sentencing judge in any of the cases. It is material of a general character relevant to sentencing which can be, and sometimes is, put before a sentencing judge. This material was said by the Crown to be relevant to the determination of the individual cases, as well as to the formulation of sentencing guidelines of general application.
                50 The Crown also put before the Court sentencing statistics for s97(1) as compiled by the Judicial Commission of New South Wales. No objection was taken to the Court referring to this material, either in the individual cases or for purposes of the formulation of guidelines. Such information is, of course, available to all sentencing judges and, accordingly, this Court can act on the basis that the trial judge either had, or could have had, access to it if he or she wished. The Court of Criminal Appeal must, of course, be in the same position in this regard. All parties conceded that access to this information was appropriate.
                51 The Respondents to the Crown appeals and the Appellant in the severity appeal objected to the Court taking into account the additional material in the individual cases. This material was of three kinds:
                    (i) Crime statistics on the incidence of armed robbery offences, including police data, court data and victim surveys.
                    (ii) Academic literature on the effects of armed robbery on victims, including both physical and psychological effects.
                    (iii) A report by a clinical psychologist, prepared for purposes of this hearing, on the psychological impact of armed robbery upon victims.
                52 The objection to the use of this evidence for the individual cases was based on the proviso to s12(1) of the Criminal Appeal Act 1912. That section is in the following terms:
                “(1) The court may, if it thinks it necessary or expedient in the interests of justice:
                    (a) order the production of any document, exhibit, or other thing connected with the proceedings, and
                        (b) order any persons who would have been compellable witnesses at the trial to attend and be examined before the court, whether they were or were not called at the trial, or order any such persons to be examined before any judge of the court or before any officer of the court, or justice, or other person appointed by the court for the purpose, and admit any deposition so taken as evidence, and
                        (c) receive the evidence, if tendered, of any witness (including the appellant) who is a competent, but not a compellable witness, and
                        (d) where any question arising on the appeal involves prolonged examination of documents or accounts, or any scientific or local investigation, which cannot, in the opinion of the court, be conveniently conducted before the court, the court or any judge thereof may refer the question for inquiry and report to a commissioner appointed by the court, and act upon the report of any such commissioner so far as the court thinks fit, and
                        (e) appoint any person with special expert knowledge to act as assessor to the court in any case in which it appears to the court that such special knowledge is required for the determination of the case;
                        and exercise in relation to the proceedings of the court any other powers which may for the time being be exercised by the Supreme Court on appeals or applications in civil matters, and issue any warrant or other process necessary for enforcing the orders or sentences of the court: Provided that in no case shall any sentence be increased by reason of, or in consideration of any evidence that was not given at the trial. (Emphasis added)
                53 Section 12 was part of the original Criminal Appeal Act when it was enacted in 1912. The Act as at that time did not make any provision for appeals by the Crown. By s6(3) of the Act as originally enacted, this Court was expressly empowered to increase the sentence on an appeal by a convicted person. This necessitated the proviso in s12.
                54 This is the explanation for the proviso, with which I agree, given by McPherson J in Beldan (1986) 21 ACrimR 159 at 167, where the Court was concerned with the equivalent provision in s671B of the Criminal Code (Qld), inserted in 1913. I also agree with his Honour’s conclusion that the proviso to s12(1) applies to Crown appeals, even though they were subsequently introduced.
                55 In Behar NSWCCA 14 October 1998, I made reference to the possible difficulty which s12 may present for the preparation of guideline judgments. Behar was handed down a few days after Jurisic. Subsequently the Parliament amended the Criminal Procedure Act 1986 with the introduction of Pt 8 “Sentencing Guidelines” by the Criminal Procedure Amendment (Sentencing Guidelines) Act 1998. Section 26 of the Act now empowers the Attorney General to apply to the Court of Criminal Appeal for a guideline judgment.
                56 Section 29A now provides:
                “(1) Nothing in section 12 of the Criminal Appeal Act 1912 limits the evidence or other matters that the Court may take into consideration in giving a guideline judgment (whether or not on application under section 26) and the Court may inform itself as it sees fit.
                    (2) The Court must not increase a sentence in any appeal by reason of, or in consideration of, any evidence that is used by the Court in giving a guideline judgment in the appeal and that was not given at the trial.”
                57 By reason of the words “whether or not on application under s26” in 29A(1), the material on which the Crown sought to rely, can be taken into account by this Court in preparing a guideline judgment with effect in futuro. This was conceded by all parties.
                58 Section 29A puts beyond doubt, in the context of the promulgation of sentencing guidelines, what would probably have been accepted even without legislative provision. The formulation of guidelines, even with respect to the exercise of statutory discretions, often requires information which goes beyond the circumstances of an individual case.
                59 The recognition that judges do in fact develop and change the law, as distinct from merely declaring the law, has received general acceptance comparatively recently. Accordingly, the circumstances in which the Court will receive material, including evidence in the formal sense, for purposes of such policy development have never been clearly articulated and established. (See Davis “The Judiciary - Maintaining the Balance” in Finn (ed) Essays on Law and Government vol 1 pp283-285; Doyle “Implications of Judicial Law Making” in Saunders (ed) Courts of Final Jurisdiction: The Mason Court in Australia esp at pp96-97). It is plain that cases which raise policy issues cannot be allowed to become law reform commission inquiries.
                60 The means of acquiring information for the purposes of policy development should not be confined by the rules of evidence developed for fact finding with respect to matters that only concern the parties to a particular case. (See Davis “An Approach to Problems of Evidence in the Administrative Process” (1942) 55 HarvLRev 364 esp at 402; Davis “Judicial Notice” (1955) 55 Col. LR 945 esp at 952-953; Carter “Judicial Notice” in Campbell and Woller (eds) Well and Truly Tried (1982) at pp92-94; Ligertwood Australian Evidence (3rd ed, 1998) par 6.42).
                61 Similar issues have arisen with respect to the determination of constitutional facts. As Brennan J has put it:
                “There is a distinction between a judicial finding of a fact in issue between parties upon which a law operates to establish or deny a right or liability and a judicial determination of the validity or scope of a law when its validity or scope turns on a matter of fact. When a court, in ascertaining the validity or scope of a law, considers matters of fact, it is not bound to reach its decision in the same way as it does when it tries an issue of fact between a parties. The validity and scope of a law cannot be made to depend on the course of private litigation.” (Gerhardy v Brown (1984) 159 CLR at 141-142).
                62 His Honour went on to refer to Breen v Sneddon (1961) 106 CLR at 406-406 and Commonwealth Freighters Pty Ltd & Boland v Sneddon (1959) 102 CLR 280 at 292 and added:
                “The court may, of course, invite and receive assistance from the parties to ascertain the statutory facts, but it is free also to inform itself from other sources. Perhaps those sources should be public or authoritative, and perhaps the parties should be at liberty to supplement or controvert any factual material on which the court may propose to rely, but these matters of procedure can wait consideration on another day. The court must ascertain the statutory facts ‘as best it can’ and it is difficult and undesirable to impose an a priori restraint on the performance of that duty.”
                63 This reasoning is applicable to the determination of sentencing principles or the promulgation of a guideline judgment operative in futuro.
                64 Different considerations apply in the sentencing context, when material relevant at a policy level is sought to be relied upon, either directly or indirectly, by the application of the new or differently articulated policy, principle or guideline, to the case on appeal. Issues of double jeopardy and disparity arise in this context.
                65 The Respondents to the Crown appeal, and the Appellant in the severity appeal, rely on the proviso to s12(1) of the Criminal Appeal Act, as reinforced by s29A(2) of the Criminal Procedure Act 1986. These provisions reflect the principle of double jeopardy. The Respondents submit that none of the material should be taken into account with respect to the particular cases before the Court.
                66 No submission was made that either the criminal statistics or the academic literature could not have been taken into account by the sentencing judge. The focus of the submissions was on the double jeopardy element, reflected in the two statutory provisions. Accordingly, it is unnecessary to consider whether Re Richardson [1920] SALR 24, which turns on the “generally known” formulation from Holland v Jones (1917) 23 CLR 149, is still good law in the light of the different formulation in s144 of the Evidence Act 1995.
                67 Nor was any submission made that, in any case, there was an inadequate opportunity to deal with the additional material (cf s144(4) Evidence Act 1995).
                68 The Crown eventually conceded that the report of the clinical psychologist expressly prepared for use in these proceedings, cannot be relied on in the individual cases. This leaves two categories of material:
                    (i) Crime statistics
                    (ii) Academic literature on victim impact.
                69 The issue turns on the proper construction of the word “evidence” in each of the proviso to s12(1) of the Criminal Appeal Act and s29A(2) of the Criminal Procedure Act. It is plain that in the later Act, the Parliament intended to use the word in the same sense as it was used in the proviso to the former Act.
                70 It is noteworthy that there is a distinction between subs (1) and subs (2) of s29A of the Criminal Procedure Act in this regard. Subsection (1) refers both to “evidence” and “other matters”, whereas subs (2) refers only to “evidence”.
                71 The principal thrust of the submissions for the Respondents to the Crown appeals was that the additional material on which the Crown wished to rely is “evidence” because the Crown seeks to prove as facts the following:
                    (i) The existence of an increase in the number of armed robberies in recent times.
                    (ii) The extent and intensity of the impact of armed robberies on victims.
                72 Both of these, it is said, are issues of “fact” about which the Crown seeks to adduce “evidence” in the form identified.
                73 Merely to identify an issue as one of “fact” does not lead to the conclusion that the Court is restricted to “evidence” with respect to its determination. For example, the meaning of an ordinary English word has long been held to be a matter of fact not law, yet the Court consults dictionaries and other books in order to determine the meaning. (See Australian Gaslight Company v The Valuer General (1940) 40 SR (NSW) 126 at 137).
                74 An appellate court may take judicial notice on appeal of something which has not been “noticed” at first instance. (See Warren v Pilkington (1960) Tas SR 6 at 9-15). This must, of course, be subject to any contrary statutory provisions.
                75 The Court may “notice” certain notorious facts of which ordinary persons are presumed to be aware. The Court can “notice” such facts either directly, or after being “reminded” of such facts by appropriate information or evidence. (See eg Holland v Jones (1917) 23 CLR 149 at 153). The Court may make reference to works of reference or authority in order to “remind it” of what it “knows”. (See McQuaker v Goddard (1941) KB 687 at 700-701). Sometimes evidence will be required.
                76 The information which the Court acquires in this way is not “evidence strictly so called”. (See McQuaker v Goddard supra at 700; Baldwin & Francis Limited v Patents Appeal Tribunal [1959] AC 663 at 691; Saul v Menon (1982) NSWLR 314 at 325; Wigmore on Evidence (3rd ed) par 2568a; Stone & Wells Evidence: Its History and Policies (1991) p174; Noakes “The Limits of Judicial Notice” (1958) 74 LQR 59 at 63). Nor, in my opinion, is “knowledge”, of which proof is not required under s144 of the Evidence Act, “evidence strictly so called”.
                77 The proviso to s12, and its interstate equivalents have been the subject of judicial consideration. In Beldan supra, the Queensland Supreme Court held that the proviso did not preclude the receipt by the appellate court of evidence to show that the sentencing process was affected by deception on behalf of the Respondents to a Crown appeal (Beldan supra at 167 and 178). This Court has applied Beldan and admitted evidence to show that the sentencing judge was deceived by evidence of a promise to give assistance to the Crown which, this Court found, the Respondent had no intention to honour. (See Chanh Nghia Ly NSWCCA 16 December 1992 at 10-11. See also Todhunter v Attorney-General (Com) (1994) 52 FCR 228 at 246).
                78 The issue also arose in J (1992) 64 ACrimR 441, where Olsson J, with whom King CJ and Mulligan J agreed, held that the South Australian equivalent section (s359 of the Criminal Law Consolidation Act 1925 (SA)) did not prevent the receipt of evidence as to the conduct of the Respondent to a Crown appeal in failing to honour an undertaking to provide assistance to the Crown in other proceedings. His Honour said:
                “There is no doubt that the clear intention of the Parliament was to ensure that, so far as possible, an offender was not unfairly placed in double jeopardy; and that, where further evidence not given at trial is admitted on appeal, that ought not, in the normal course, lead to an increase in the sentence imposed.

                    When the proviso to s359 speaks of evidence that was not given at the trial it is, in my opinion, directing its attention to matters going to either the commission of the offence or the personal circumstances of the accused. It is not concerning itself with the situation in which the Crown is merely informing the court that the person sentenced has failed to honour a specific undertaking given to the court as an express basis upon which the sentence is imposed.” (459)
                79 In Veen v The Queen (No 2) (1987-88) 164 CLR 465, one of the issues before the Court was whether the intermediate Court of Appeal should have received material about the state of psychiatric services in New South Wales gaols as facts relevant to the determination of sentencing principles. In the event the Court held that it was unnecessary to do so but added, by way of obiter dictum:
                “It is unnecessary to consider the procedure adopted by the Court of Criminal Appeal in admitting evidence as to the state of gaol psychiatric services but, subject to appropriate procedural safeguards, we would not regard that court as being precluded from receiving further material relevant to the adoption of sentencing principles or sentencing policy of general application.” (473)
                80 Notwithstanding the reference to “policy of general application”, the issue in Veen (No 2) concerned the use of any such “policy” in the particular case before the court.
                81 This passage from Veen (No 2) was applied by this Court in Hallocoglu (1991) 29 NSWLR 67, where the Court received a document, over objection, being the booklet supplied to prisoners serving periodic detention outlining the operations of that scheme. The Court expressly rejected the submission that such material constituted “evidence” within the meaning of s12(1) of the Criminal Appeal Act. Hunt CJ at CL said:
                “That submission must be rejected … We are entitled to inform ourselves from such material as to the true consequences of an order that a sentence be served by way of periodic detention.” (73-74)
                82 It may be that when Hunt CJ at CL held in Hallocoglu that s12 did not prevent the Court informing itself about the administration of the periodic detention system, he had in mind a test for “evidence” similar to that propounded by Olsson J in J: “matters going to the commission of the offence or the personal circumstances of the accused”. Whilst Hallocoglu was, in certain respects, a guideline judgment, the use of this material was not restricted to the formulation of sentencing principles. The Court had reference to the material for the purposes of the individual case.
                83 I am unable to distinguish the criminal statistics on which the Crown seeks to rely in this case, from the material relied upon by the Court in Hallocoglu.
                84 In my opinion, it should now be accepted that “evidence” in s12 of the Criminal Appeal Act and s29A of the Criminal Procedure Act is confined to “matters going to the commission of the offence or the personal circumstances of the accused”. On this basis the criminal statistics are not “evidence” within the proviso.
                85 The academic literature on the effects on victims is in a different position. Each individual case had its own victim or victims. The objective gravity of each case is determined by, inter alia, the effects on victims in the particular case. The general literature is only of assistance at the level of sentencing principle. Whilst it may not be “evidence” within the proviso to s12, it is not material to which the Court should have regard in the individual cases.
                Crime Statistics and Need for Guidelines
                86 The Crown sought to rely on crime statistics for the purpose of indicating that sentencing guidelines were required by reason of the prevalence of the offence. Prevalence is a relevant consideration when deciding an appropriate level of sentence. (See eg Giles v Barnes (1969) SASR 174 at 181; Peterson (1983) 11 ACrimR 164 at 167-168, 169; Henderson (Jana Brian) (1991) 58 ACrimR 369 at 373; Cuthbert (1967) 86 WN (pt 1) NSW 272 at 278; Colman (Anthony Keith) NSWCCA 5 November 1997; Kukunoski NSWCCA 17 August 1989; Retton (Timothy Craig) NSWCCA 14 November 1994). Prevalence has been acknowledged as a reason for establishing a guideline in English guideline judgments. (See eg Cunningham (1993) 96 CrAppR 422 at 425; Brewster (1998) 1 CrAppR 220 at 224 per Lord Bingham).
                87 “Prevalence” may refer to a situation in which a particular crime occurs with such frequency that it has a salience beyond the persons immediately affected by the crime and, accordingly, impacts on society by changing patterns of behaviour out of a sense of apprehension. None of the material presented by the Crown to this Court suggests that armed robbery is “prevalent” in this sense. Indeed the Crown submissions referred to armed robbery as “a relatively rare crime”.
                88 Rather, what the Crown sought to prove was that the incidence of this particular crime had increased over recent years, so that the objective of general deterrence was entitled to greater weight than it might have hitherto received.
                89 Three sources of data were presented to the Court: police statistics, court statistics and victim surveys. By reason of changes in the manner in which recorded crime statistics were kept by the NSW Bureau of Crime Statistics and Research over the years, police statistics do not permit long term trends to be analysed. The data presented to this Court was broken up into four periods: 1982-1989, 1991-1993, 1994-1995 and 1995-1997. Of the statistics for these respective periods none, save the last three year period between 1995-1997, suggests any changes of statistical significance.
                90 The Crown relied on the fact that for the period 1995-1997 there was a significant increase in the number of armed robberies involving a weapon other than a firearm. The rate per 100,000 for such offences increased from 30.5 in 1996 to 53.9 in 1997 and reached 69.2 in 1998. (NSW Bureau of Crime Statistics and Research NSW Recorded Crime Statistics 1998 p41). It is unlikely that, in the case of this offence, such an increase could have been caused by an increased propensity to report the commission of offences.
                91 The second source of data was the NSW Criminal Court statistics which covered the period 1991-1997, during which there does not appear to be any significant change. However these figures do show fluctuations over the years, both up and down.
                92 The third source of statistics relied upon by the Crown was victim surveys. This information indicated that over the period 1990-1997 there was no evidence of any systematic upward or downward trend with respect to robbery victimisation.
                93 In the light of the other material, the reported crime statistics for the years 1995-1997, which suggest a significant increase over that period, should be treated by this Court with caution when it is suggested that changes in the incidence of the crime are such as to require a response in the sentencing practice of the courts. Nevertheless, the increase over the three year period is entitled to weight when deciding whether or not a guideline judgment is appropriate and, if so, what change with respect to past sentencing practice is appropriate.
                Victim Impact and Need for Guidelines
                94 The second body of material to which the Crown sought to make reference concerned the impact of armed robbery on victims. This included academic literature and the report of a clinical psychologist prepared for these proceedings. This material was said to establish the objective gravity and seriousness of the offence of armed robbery. There is no doubt that impact on victims is an aspect of the seriousness of an individual offence. General patterns of impact of the character referred to in the literature to which the Crown referred, including statistical surveys, confirm the seriousness of the offence.
                95 Plainly the actual impact in each particular case will vary and, appropriately, cause variations in the sentence imposed. This is not a manifestation of inconsistency. Rather, it represents the consistent application of a principle which varies in its import according to the circumstances.
                96 The surveys and other literature, to which the Crown referred, establish that armed robbery is perceived as a life threatening situation for a majority of victims. In one Australian survey 86 percent reported the robbery as being the most threatening experience they had ever had. Victims experienced both physiological and psychological problems. The physical effects included chronic nervousness, insomnia, nightmares, headaches, digestive problems. Psychological problems included generalised fear, depression, aggressiveness, mood changes. Sometimes the response was such as to develop into the clinical condition of post traumatic stress disorder.
                97 The surveys and academic literature confirm what trial judges and appellate judges would in any event know, namely, that armed robbery is a serious offence by reason, inter alia, of the significant impact it has on its victims. Indeed, the submissions on behalf of the Respondents to the Crown appeals did not, in contrast with criticism of the reliance by the Crown on crime statistics, challenge the materials put before the Court on the impact of the crime of armed robbery on its victims.
                98 The Crown also put before the Court a report prepared for purposes of this appeal by a clinical psychologist on the effects of emotional distress in the aftermath of personal trauma, not limited to cases of armed robbery. Nothing in the report was startling or controversial and no submissions were directed to it.
                99 This material confirms what this Court would in any event accept on the basis of commonsense and common knowledge. Armed robbery is not simply a crime against property. It is a crime against persons. Furthermore, the fear engendered by the perpetrator of this crime, together with the continued adverse effects on its victims, establish armed robbery to be a serious crime which requires condign punishment.
                Sentencing Statistics and Need for Guidelines
                100 The primary material before the court for the purpose of determining whether or not a guideline judgment is appropriate is to be found in the statistics kept by the Judicial Commission of New South Wales. Those statistics are available for all sentences imposed between December 1994 and October 1998 for offences against s97(1) of the Act, which extends to both armed robbery and robbery in company. In the period of almost four years covered by the statistics, there were a total of 835 cases. This is a significant number from which general conclusions as to sentencing practices can be drawn.
                101 The most important single aspect of the statistics is that of the 835 cases, a total of 688 (eighty-two percent) resulted in a full-time custodial sentence. That is to say, 147 cases, eighteen percent of the total, did not result in a full-time custodial sentence. Of those, the largest number, 64 (eight percent) resulted in a sentence of periodic detention and 55 (seven percent) resulted in a community service order.
                102 The Sentencing Information System of the Judicial Commission provides the following information for the 688 cases in which a sentence of imprisonment was imposed, divided into charts for the Full Terms and those for the Minimum or Fixed Terms.
                103

                At the other end of the spectrum, only seven individuals received more than half the maximum term of twenty years, i.e. more than ten years, as full terms. Furthermore, only about one-quarter of the entire sample of 688, being 163 cases, received full sentences of more than one-quarter of the maximum, i.e. more than five years.
                104 The chart with respect to the minimum or fixed terms is also revealing. The statistics do not allow the removal of the cases of fixed terms in order to look at the sentencing practice with respect to the relationship between minimum and full terms. On the basis of experience, it can be said that fixed terms are comparatively few in number. The statutory formula in the Sentencing Act 1989 (NSW) is for an additional term not exceeding one-third of the minimum term, unless “special circumstances” are found to exist.
                105 There are 307 cases out of the 688 total (45%) for which a minimum (or fixed) term of eighteen months or less was imposed. The application of the statutory ratio suggests that something of the order of that number of cases should have received full terms of twenty-four months or less (without making express allowance for the number of fixed terms). However, the full terms chart shows that only 73 individuals (not about 307) had sentences of twenty-four months or less imposed.
                106 For the next bracket up, the figure for minimum terms is 426 individuals who received two years or less for which the statutory ratio would imply a full term of thirty-two months. The Judicial Commission’s statistics provide full terms for thirty months and thirty-six months. In the former case, there were only 106 individuals sentenced and, in the latter case, 246.
                107 These statistics indicate that the “special circumstances”, which justify a ratio other than the statutory ratio, are found in the majority of cases. Furthermore, the figures strongly suggest that it is very often the case, as suggested in the particular instances before the court on this occasion, that trial judges have imposed a minimum term of significantly less than half of the full term.
                108 This is an offence for which the maximum penalty is twenty years. The statistics show that of the 688 persons sentenced to prison, sixty-two percent of the total receive an actual prison term (minimum or fixed) of two years or less. Eighty percent receive actual minimum or fixed terms of three years or less.
                109 These statistics indicate the following conclusions as to sentencing practices in the District Court for offences against s97(1) of the Crimes Act:
                (i) Non-custodial sentences are not confined to exceptional cases. The fact that 147 individuals received non-custodial sentences out of a total of 835 (i.e. 18%) is not consistent with an “exceptional case” test.
                (ii) The leniency of the sentencing pattern is also suggested in the full terms. (Over 60% of those who were in fact sentenced to prison received one fifth or less than the maximum, i.e. four years or less).
                (iii) The pattern of leniency is reinforced by the extent to which the statutory ratio between minimum and fixed terms has been altered so that over 60% of those who do receive prison sentences, receive a minimum or fixed term of two years or less.
                (iv) The large proportion of lenient sentences is also suggestive of inconsistency in sentencing practices between the group who receive full terms of 3-4 years or thereabouts and the group below that.
                110 These statistics strongly suggest both inconsistency in sentencing practice and systematic excessive leniency in the level of sentences. They justify the promulgation of a guideline judgment.
                    CCA Cases
                111 One of the factors considered by the Court in Jurisic to be relevant to the determination of a need for a guideline judgment was the history of Crown appeals against sentence for the offence there under consideration, s52(A) of the Crimes Act. This Court had said, on a number of occasions, that that offence had to be treated as being more serious than it had been regarded. Notwithstanding those indications, there was a continued flow of, almost invariably successful, Crown appeals.
                112 The position with respect to s97(1) in this Court is not as unequivocal as the history with respect to the offence under consideration in Jurisic. Nevertheless, some aspects of this Court’s prior consideration of Crown appeals from sentences for s97(1) convictions are relevant.
                113 First, and most important, is the frequency with which the Court has stated that a non-custodial sentence for this offence could only be imposed in exceptional circumstances. As Hunt CJ at CL said in Roberts (1994) 73 A Crim R 306 at 308:
                “This Court has always made it clear that armed robbery is to be regarded as an offence of the utmost gravity which, except in the most exceptional circumstances, must carry a full time custodial sentence: Murray (unreported, 11 September 1986) at p5; Kingsbeer (unreported, 29 July 1988) at p7; Valentini (1989) 46 A Crim R 23 at 26; Readman (1990) 47 A Crim R 181 at 104-105; Diamond (unreported, 18 February 1993) at p2; Hetherington (unreported, 25 February 1993) at p4; Maddocks (unreported, 25 November 1993) at pp2, 6. That view has been said by this Court to be a sentencing principle resulting from its considered decisions to which sentencing judges should not merely pay lip service, Maddocks (at p6). Wood J, delivering the principal judgment, went on to say:
                    ‘I would repeat, although I wonder why it is necessary that I should do so, that it is only in the most exceptional circumstances that anything other than a non-custodial sentence should be imposed for armed robbery. Necessarily, there will be cases which constitute an exception but they will be few and far between.’
                        With those observations, I express my complete agreement. I prefer the phrase “most exceptional circumstances” which he used to the phrase “wholly exceptional and unusual circumstances” subsequently employed in Crotty (unreported, NSWCCA 28 February 1994) at p5.”
                114 The cases referred to by Hunt CJ at CL indicate the frequency with which this Court had reiterated this basic proposition prior to 1994.
                115 The proposition, as summarised and reiterated in Roberts, has been applied subsequently in successful Crown appeals (Wright (1997) 93 ACrimR 48; Bragias NSWCCA 12 March 1997; Khoury NSWCCA 5 December 1994; Amohanga NSWCCA 25 May 1995).
                116 It has also been referred to in Crown appeals in which the Court has found the first instance sentence to be manifestly inadequate but, in the exercise of its discretion, decided to dismiss the appeal. (Sharpe NSWCCA 27 September 1994; Randall NSWCCA 19 September 1994; Kerr NSWCCA 26 August 1997; Gitt NSWCCA 18 May 1998).
                117 The Court has affirmed the basic proposition on occasions on which it has held that a finding of exceptional circumstances was open to the trial judge (Jones NSWCCA 15 April 1994; Georges NSWCCA 30 May 1996; Latouff NSWCCA 12 December 1996; Tocknell NSWCCA 28 May 1998).
                118 To use the words of Wood J from Maddocks, which Hunt CJ at CL quoted with approval in Roberts, non-custodial sentences should be “few and far between”. One thing that is clear from the Judicial Commission’s statistics is that non-custodial sentences imposed by trial courts for this offence are common. They cannot be described as “few and far between”. This long line of authority has not been reflected in the sentencing practice of trial judges.
                119 The distinction between the precedents in this Court on this offence, and the precedents considered in Jurisic, is that the Court has been prepared to uphold the finding of exceptional circumstances on a number of Crown appeals, whereas in Jurisic the disposition of appeals was almost all one way. However, prior to the present hearing, the pattern of sentencing by trial judges and this Court’s precedents on appeal, have never been reviewed in a systematic way. It is one of the advantages of a system of guideline judgments that this Court can review its own prior decisions from the perspective of ensuring consistency in the guidance it gives to trial judges. The pressures on the Court do not necessarily permit it to review its own decisions from this perspective in the normal course.
                120 In the course of its prior consideration, the Court has had occasion to make reference to the sentencing statistics. In Tocknell Hulme J, with whom Hidden J and Carruthers AJ agreed, said, with reference to the conclusion that a finding of exceptional circumstances was within the sentencing judge’s discretion:
                “In reaching that conclusion I place no weight on the sentencing statistics prepared by the Judicial Commission which were put before this Court. Insofar as some of those statistics suggest that something in the order of forty-seven percent of offenders guilty of the offences with which the respondent was charged and under twenty-one, had not been sentenced to full time custodial sentences, those statistics suggest to me that the judges of the District Court are not paying proper regard to the principles which this Court has laid down. If that impression be correct, and if the tendency continues, it may be that the traditional approach of this Court to appeals by the Crown may need to be reviewed.”
                121 The issue arose in a more acute form in the case of Ross (NSWCCA 14 May 1997). In that case, the sentence of a two year minimum term and a two year additional term was within the trial judge’s discretion to find “the case to be an exceptional one deserving leniency”. Studdert J and Levine J both agreed with the result, but with the following observations:
                “… The success of the appeal depended upon the Crown establishing that in all the circumstances the sentence was manifestly inadequate. In this regard the Crown case was not assisted by sentencing statistics produced to the court. Those statistics did not illustrate that the sentence here imposed was outside the range. Of course caution has to be used in resorting to such statistics and an appeal cannot be determined simply by reference to statistics of the nature it provided. …However, I am not persuaded that this Court should intervene.”
                122 Considerations of this character are plainly material. Particular scrutiny is appropriate where statistics indicate that the sentence under appeal is at the top or bottom of the range of sentences actually imposed, in circumstances where its objective characteristics did not indicate that it is appropriately so characterised. (See e.g. Bugmy (1991) 69 CLR 525 at 538; Bloomfield (1998) 44 NSWLR 734 at 739F). However, the court in Ross did not have before it a challenge to the usual level of sentencing in the way that that issue has been presented to the Court on this occasion. The judgments in Tocknell and Ross are precursors to the present judgment.
                123 The number of cases in which non-custodial sentences have been imposed indicates that the long established principle that such must be confined to “exceptional circumstances” has not been implemented by sentencing judges. The position is not quite as clear with respect to the level of the sentencing where custodial sentences are imposed. It is overwhelmingly likely that the approach to sentencing reflected in the imposition of non-custodial sentences has also been reflected in the length of custodial sentences when imposed. The sentencing statistics suggest that that is so.
                124 In Waldron (NSWCCA 3 March 1994) Hunt CJ at CL said (at p3):
                “This Court has on numerous occasions said that the range of sentences imposed for armed robberies, particularly serious ones, should be more deterrent than those which are in fact being imposed by the sentencing judges; see, for example Flack (12 December 1989); Smith (12 December 1984); Petrinovic (18 September 1990); Va (11 November 1993).” (3)
                    See also Ellis (1993) 68 ACrimR 449 at 462; Lawson NSWCCA 12 December 1997.
                125 The sentencing statistics hitherto set out, indicate that the views expressed in Waldron have not been implemented. In part, this may have occurred because of the number of cases of findings of “exceptional circumstances” which this Court has endorsed on a basis other than the exercise of its discretion. A guideline judgment is required to ensure that the pattern of leniency and inconsistency indicated in the sentencing statistics is discontinued.
                Guideline Precedents
                126 The determination of an appropriate sentencing guideline for offences against s97(1) must commence with a recognition that the objective and subjective factors relevant to the exercise of the sentencing discretion, may vary over a wide range. See eg Brown (1989) 17 NSWLR 472 at 473-474. As I said in Jurisic with respect to s52A:
                “The nature of the offence is not such that the Court can devise a simple table in which indicative penalties are linked to a quantitative measure of the offence … What can be done, however, in case of an offence concerning a wide range of conduct which varies qualitatively rather than quantitatively, is to indicate in a general way the kind of case which would usually require a particular kind or level of sentence, whilst acknowledging that there will always be exceptional cases.” (35-36)
                127 A consideration of the practice of other criminal appellate courts, with respect to guidance for sentencing for armed robbery, suggests that such an approach is appropriate for this offence. I commence with the English cases.
                128 In 1972, the Court of Appeal (Criminal Division) heard together a number of severity appeals from convictions for armed robberies involving banks. The Court identified a “starting point” of fifteen years imprisonment for what the Court described as a “normal” bank robbery, where firearms were carried and no serious injury done. (Turner (1975) 61 CrAppR 67 at 91).
                129 In Gould (1983) 5 CrAppR(S) 72, the Court considered a number of cases of robbery of small business premises where the offenders were armed with guns, imitation guns, or knives. The Court confirmed the guideline in Turner as applicable to a serious type of case but added:
                “There are so many possible combinations of circumstance that it is difficult to give any precise indication of the so-called normal sentence for any particular type of robbery.” (75)
                130 The Court went on to identify a list of mitigating and aggravating circumstances. The Court found that the sentences in the three appeals before it (between five and eight years) were not excessive. It did not specify a starting point or sentencing range.
                131 The Court continued with this approach when it dealt with a number of Crown appeals in Attorney-General’s References Nos 3, 4, 8, 9, 10, 11 and 16 of 1990 (Dickson & Ors) (1991) 92 CrAppR 166. All these cases involved small shops. The Court allowed the appeals and substituted verdicts of between three and a half years and six years.
                132 The Court has subsequently referred to a seven year sentence in cases of this character as if it were a starting point. The case involved a very small bank branch and an imitation gun (Attorney-General’s Reference No 14 of 1991 (Reed) (1992) 13 CrAppR (S) 446). Lord Lane CJ said:
                “… it is very seldom that in these circumstances a sentence of much less than seven years would be appropriate, the purpose of the sentence being threefold: first of all to deter the offender from behaving in this way again; secondly to deter others from arming themselves with weapons and holding up banks; and thirdly, a matter which is sometimes overlooked, to punish the offender for having carried out this wicked crime and having put several people in fear of death.” (449)
                133 The seven year sentence in Reed was referred to as a starting point by Lord Taylor CJ in Attorney-General’s Reference No 7 of 1992 (Khan) (1993) 14 CrAppR (S) 122 at 125, 126, which also involved a small shop and an imitation pistol. His Lordship added:
                “It has to be realised that that type of shop is very often staffed by only one person who may be unable to defend himself or herself. It is unlikely that there will be any sophisticated security there, and it is a primary target for someone who wants to enrich himself quickly and successfully. It is therefore very important that the courts should indicate by the sentences passed that that type of offence will be punished severely.” (126)
                134 I should note that the Court of Appeal has subsequently issued a separate guideline judgment with respect to the use of firearms in a variety of offences (Avis (1998) 1 CrAppR 420).
                135 This reluctance in Gould to specify a range or starting point was based on the wide variation in the factual circumstances of individual offences. The reluctance to offer any guidance, where complete guidance is not feasible, has been criticised. (Ashworth “Techniques of Guidance on Sentencing” (1984) CrimLR 519 at 529-530; Ashworth Sentencing and Criminal Justice (2nd ed, 1995) at 30. It may be that that subsequent decisions are a response to this criticism.
                136 Guidance may be given by identifying certain commonly recurring categories of a specific offence. The New Zealand Court of Appeal has done this in Moananui (1983) NZLR 537. Guideline judgments in New Zealand are “bottom up” not “top down” guidelines, i.e. they purport to describe, rather than prescribe, sentencing practice. (See Hall “Reducing Disparity by Judicial Regulation: Sentencing Factors and Guideline Judgments” (1991) 14 NZ ULR 208 esp at 223-224).
                137 In Moananui the Court identified three categories:
                    (i) Planned armed robbery at premises such as banks which endanger the safety of considerable numbers of people: which usually attract sentences of 6-8 years.
                    (ii) armed robbery involving intrusion into dwelling houses, usually at night: examples given range from 4-6 years.
                    (iii) armed robberies of smaller premises involving less extensive risk of injury and smaller sums of money: the range of sentences is lower. (Examples range from 2-5 years).
                138 A more overtly structured approach has been adopted in Hong Kong. In Mo Kwong Sang (1981) HKLR 610, the Court identified what it described as an “ordinary case” to be a starting point: armed robbery by an offender carrying a knife or other dangerous weapon, which was displayed to the victim. The Court said that an “appropriate” sentence for this case was five years. It indicated appropriate levels for two other common factual variations: invasion of private premises (six years) and physical violence (seven years).
                139 The Court also identified a list of aggravating and mitigating factors, noted that use of firearms would require more severe sentences, and affirmed the discretion of the sentencing judge to take into account subjective circumstances. The nature of the guidance provided by the appellate court was to:
                “… indicate the level of sentence for armed robbery with which we would not interfere. (611)
                140 I have noted above that Canadian criminal appellate courts have adopted a “starting point” approach to providing guidance to trial judges. This involves the identification of a typical case and the application of aggravating and mitigating factors. Courts in Alberta, Nova Scotia and New Brunswick have applied this approach to armed robbery. (See Young The Role of an Appellate Court in Developing Sentencing Guidelines Canadian Sentencing Commission (1988) p21ff).
                141 In Nova Scotia the appellate court has lain down a three year minimum sentence (Brennan and Jensen (1975) 11 NSR (2d) 84 at 88; Hingley (1977) NSR (2d) 541 at 544). The New Brunswick Court of Appeal has referred to three years as a normal sentence (Chaisson (1975) 24 CCC (2d) 159). In both cases the sentencing level was identified as appropriate to a young first offender, but there was little in the way of precision about other aspects of the offence or of the offender.
                142 In Johnas (1982) 2 CCC (3d) 490 the Alberta Court of Appeal focused on cases of robberies of “small commercial establishments, open at night for service to and convenience of the public, where often a single person is in charge” (494). The objective circumstances emphasised the degree of vulnerability. The category identified involved a threat of violence, but no actual violence. The Court identified the category as follows:
                “… what should be regarded as a fit sentence in Alberta for unsophisticated armed robbery of unprotected commercial outlets in the absence of actual physical harm to the victim and with modest or no success.” (495)
                143 The Court determined three years as a “starting point” and added:
                “… judicial reasoning as to a fit sentence for any offence must start with a norm for the type of offence involved. That norm is arrived at by comparisons to other cases, by experience, by the seriousness of the offence and by its prevalence. Having determined that norm, the court will look at the factors of mitigation and aggravation. Specific cases are not to be treated as precedents. Each is a unique mixture of aggravating and mitigating factors.” (499)
                144 In subsequent cases the Alberta Court of Appeal has specified five years as a starting point for bank robberies (Kurich (1982) 9 WCB 138) and four years for night deposit robberies (Hall (1983) 10 WCB 138).
                145 The Court of Criminal Appeal of Western Australia has, with respect to armed robbery, provided sentencing guidance in, what I have called above, a “bottom up” fashion: i.e. it is derived from the range of sentences actually imposed by trial judges. This involves identifying a tariff rather than fixing a starting point or sentencing range. It has been called “tariff sentencing”. (Thomas Principles of Sentencing (2nd ed, 1979) p29).
                146 In 1989, what the Court described as a “conventional armed robbery of a bank or similar premises”, attracted a sentence of “five to seven years or upwards” (Norman WACCA 1 February 1989). By 1997 these levels had been increased. The Court said:
                “… sentences have tended to firm up as a result of the increasing prevalence of the offence by giving greater weight to the requirements of deterrence and less weight to the antecedents and other matters personal to the offender … The offence of armed robbery has become significantly more prevalent since 1989 and sentences have firmed up in that period. At present, the range of sentences commonly imposed for a single offence, depending on the circumstances, would be from six to nine years.” (Miles (1997) 17 WAR 518 at 521 and 522).
                147 This level applies before a discount for a plea of guilty. The Court went on to note that where a blood filled syringe was used as a weapon, an additional one year’s imprisonment should normally be imposed, because of the additional impact on the victim (523-524).
                148 The Western Australia Court of Criminal Appeal has affirmed the range of 6-9 years in subsequent decisions (Moody 12 February 1998; Moulds 4 March 1998; Jeffree 1 May 1998). In Jeffree, the Court further clarified the application of the range which was described in Miles as “commonly imposed”:
                “Offences of this type commonly do not progress beyond a threat or violence of a minor kind. Those which do could be characterised as amongst the worst type of case and would be outside the usual range.”
                149 The range of sentences for this offence in Western Australia - six to nine years - must be understood in the light of a discount for a guilty plea (said to be 20-35%, Miles at 521) and the statutory framework for sentencing, which differs from that in New South Wales. By s93 of the Sentencing Act 1995 (WA), the range of six to nine years could permit release on parole after, in broad terms, two to four years. This, again in broad terms, would equate to a minimum term under the Sentencing Act 1989 (NSW).
                150 As I have indicated above, the Full Court of the Supreme Court of South Australia has promulgated “sentencing standards” in certain cases. In Cadd (1997) 94 ACrimR 466 at 480 Doyle CJ identified the authorities which had set such standards for the offence of armed robbery. He referred to Spiero (1979) 22 SASR 543 as setting a standard and to Dube (1987) 46 SASR 118 as a case in which the Court contemplated, but refrained from, increasing the level.
                151 In Dube the Court considered a Crown submission that this standard should be increased because of the increased prevalence of the offence. The submissions made are strikingly similar to those before this Court in the seven cases now before it. King CJ noted:
                “The prosecution adduced evidence before the learned sentencing judge as to the adverse affect of armed robbery upon the victims and innocent people who become embroiled in it. There was also evidence as to the financial cost of armed robberies to financial institutions and to the community generally … The evidence led by the prosecution in this case merely reinforces the view long held on this Court that armed robbery presents a serious threat to the wellbeing of the community and that punishments imposed must reflect that fact …
                    The prosecution produced evidence before the learned sentencing judge of the increasing prevalence of this crime during the last seven years. It is unnecessary to recount the details of this evidence; it is sufficient to say that it establishes that the increase has been marked and is apparently continuing. The much discussed question of the effectiveness of imprisonment as a deterrent to crime, and in particular of the effectiveness of increased levels of punishment, was adverted to during argument. I think that it must be conceded that there is no proven correlation between the level of punishment and the incidence of crime and that there is no clear evidence that increased levels of punishment have any effect upon the prevalence of crime. Nevertheless the criminal justice system has always proceeded upon the assumption that punishment deters and that the proper response to increased prevalence of crime of a particular type is to increase the level of punishment for that crime. I think that courts have to make the assumption that the punishments which they impose operate as a deterrent. That being so, I think that it follows that the proper response, and the response which is expected by the community at large, to the increased prevalence of serious crime is increased punishment for that crime.” (119-120)
                152 The reason the Court refrained from formally increasing the standard was because of the then recently amended legislative provision on the treatment of remissions, which would have the effect of increasing sentences.
                153 Spiero involved an armed robbery of a chemist shop by three offenders. Firearms were used. The sentence of eight years fixed in that case was treated subsequently as a standard. The appropriate range has been said to be eight to twelve years, in all but “exceptional cases”. (See Fermaner (1994) 72 ACrimR 138 at 139; 61 SASR 447 at 448 based on Prendergast (1988) 147 LSJS 486 at 486-488).
                154 Fermaner was heard together with three other Crown appeals from sentences for the offence of armed robbery: Bini, Meo and Ramsden, each SACCA 21 March 1994 (unreported). The four cases are analysed in Hinton “Principled Sentencing in South Australia - Armed Robbery and the Quest for Consistency” (1995) Flinders J L.Reform 18.
                155 Three of the four cases have similar objective circumstances: robbery of a service station; offenders aged from 24 to 36 years; armed with a knife, a screwdriver and a small axe; a few hundred dollars taken; an early plea of guilty. There were differences in terms of prior record and effect on victim. The Court of Criminal Appeal set aside the sentences below and imposed sentences of five years four months (on top of five months served before sentence); with a non-parole period of four years; six years with a non-parole period of four years and 5 years 8 months, with a non-parole period of three years.
                156 In these cases (Fermaner, Bini and Ramdsen) eight years was treated as a starting point. In Fermaner the Court indicated that this would have been the appropriate head sentence but for the early plea of guilty and the period of five months in custody before sentence (140). This was reiterated in Bini and Ramsden. The plea of guilty led to a discount of two years (Hinton supra p27).
                157 Since the reaffirmation of the eight year starting point in these four 1994 cases, a system of truth in sentencing was introduced by the Statutes Amendment (Truth in Sentencing) Act 1994 (SA). As a result, the accrual of remissions, which the Court had referred to in Dube, would no longer be taken into account in sentencing. This would require a reduction of the appropriate range hitherto identified for armed robbery by the extent to which sentences had been inflated to take account of remissions. A reduction of one third would accord with the transitional provisions of the Statutes Amendment (Truth in Sentencing) Act (See Hinton supra pp25-26).
                158 This suggested a reduction in the eight to twelve years for the head sentence, to about six to eight years. This has since been affirmed by the South Australian Court of Criminal Appeal. (Drumgoon SACCA 20 November 1995; Branscherd SACCA 22 May 1996). The starting point of six years in South Australia was before a plea of guilty and other discounts (eg co-operation with the police). (See Drumgoon supra).
                159 The Queensland Court of Appeal reviewed a number of prior decisions in Hammond (1996) 92 ACrimR 450 at 456ff. The case involved pleas of guilty to three robberies of service stations by an offender armed with a steering wheel lock. After referring to the need for consistency with the prior decisions it reviewed, the Court imposed a sentence of five years with a recommendation for release on parole after two years.
                160 In a subsequent case involving a single armed robbery of a service station when armed with a knife, by an offender who pleaded guilty, the review in Hammond was referred to, the severity appeal allowed and a sentence of four years substituted. (Morton (1997) 95 ACrimR 381).
                A Guideline for New South Wales
                161 In the cases now before this Court, the Crown has propounded a starting point for New South Wales of six years full term, in circumstances where there is a plea of guilty.
                162 It appears from the cases that come to this Court, including the present proceedings, that there is a category of case which is sufficiently common for purposes of determining a guideline:
                    (i) Young offender with no or little criminal history
                    (ii) Weapon like a knife, capable of killing or inflicting serious injury
                    (iii) Limited degree of planning
                    (iv) Limited, if any, actual violence but a real threat thereof
                    (v) Victim in a vulnerable position such as a shopkeeper or taxi driver
                    (vi) Small amount taken
                    (vii) Plea of guilty, the significance of which is limited by a strong Crown case.
                163 Whilst it is possible to determine a starting point in a case of this kind, i.e. a sentence of X years imprisonment, I do not believe that the Court should do so. Rather, I propose the Court should identify a narrow sentencing range within which this Court would expect sentences in such cases to fall.
                164 There are two principal reasons why a sentencing range is appropriate for this offence:
                    (i) The seven characteristics identified above do not represent the full range of factors relevant to the sentencing exercise.
                    (ii) Many of the seven identified characteristics contain within themselves an inherent variability, eg different kinds of knives or weapons in (ii); extent of “limited actual violence” in (iv); degree of vulnerability in (v); amount in (vi).
                165 In my opinion sentences for an offence of the character identified above should generally fall between four and five years for the full term. I have arrived at this figure after drawing on the collective knowledge of the other four members of the Court with respect to sentence ranges. I have also reviewed the sentences which this Court has imposed on occasions when it has intervened, including in Crown appeals where the principle of double jeopardy applies. The proposed range is broadly consistent with this body of prior decisions in this Court.
                166 Cases consistent with the range are Sandaford 10 September 1995; Sedgwick 7 July 1992; Martin 19 February 1993; Sarkas 8 September 1992; Walters and Watkins 2 June 1994; Achurch & Brady 8 July 1994; Willett 6 November 1995; Maretta 30 October 1996; Bragias 12 March 1997; Basaga & Weleilakeba 9 August 1998.
                168 Cases above the range to varying degrees are Salameh 12 March 1991; Sneddon 28 March 1991; Marsden 15 April 1993; Antoce 22 April 1993; Walker 21 February 1994; Pettit 9 March 1994; Sneddon 28 March 1994; Nguyen 14 April 1994, Burger 19 July 1994; Rafter 23 September 1994; Harborne 12 October 1994; Murray 7 November 1995; ; Putescu 20 December 1996; White 7 April 1997; Lowe 24 April 1997.
                169 Cases below the range being sentences between 2 and 3½ years are Bateman 15 July 1993; Roberts 24 May 1994; Amohanga 25 May 1995; Davies 29 November 1995; Wright 28 February 1997; Fisher 8 September 1998; Wightman 2 November 1998.
                167 Aggravating and mitigating factors will justify a sentence below or above the range, as this Court’s prior decisions indicate. The narrow range is a starting point.
                170 In addition to factors which may arise in any case eg youth, offender’s criminal record, cooperation with authorities, guilty plea in the absence of a strong case, rehabilitation efforts, offence committed whilst on bail etc, a number of circumstances are particular to the offence of armed robbery. These include:
                    (i) Nature of the weapon
                    (ii) Vulnerability of the victim
                    (iii) Position on a scale of impulsiveness/planning
                    (iv) Intensity of threat, or actual use, of force
                    (v) Number of offenders
                    (vi) Amount taken
                    (vii) Effect on victim(s)
                171 The Respondents to the Crown Appeals also submitted that this Court should now declare that the fact that the offender was addicted to drugs, and committed the crime to obtain money to feed his or her habit, is a mitigating circumstance. This would require the Court to overturn a long line of prior decisions.
                172 In 1990, the Australian Institute of Criminology published a detailed analysis of the practice of this Court with respect to sentencing for robbery over the course of a ten year period. That report made it clear that the Court had consistently regarded drug addiction as a mere explanation, not an excuse. (Potas Sentencing Robbers in New South Wales (1990) at pp142, 163).
                173 As a joint judgment of the Court put it in Valentini (1989) 46 ACrimR 23 at 25:
                “This Court has said on countless occasions that addiction to heroin is not to be considered as a factor for the reduction of what would otherwise be an appropriate sentence for the nature of the offences which have been committed. It serves, however, to provide an explanation for the commission of the offences.”
                    This has been reiterated in a number of later cases (eg Stanford NSWCCA 23 November 1993; Atkins NSWCCA 27 May 1998).
                174 Drug addiction is one of the circumstances of a particular offence that is relevant to the sentencing exercise. It may, for example, be pertinent to the issue of impulsiveness/planning or to the weight to be given to rehabilitation in a particular case.
                175 In Engert (1995) 84 ACrimR 67, Gleeson CJ emphasised the complexity of the sentencing task and the multiplicity of the, sometimes conflicting, objectives it must serve. His Honour concluded:
                “It is therefore erroneous in principle to approach the law of sentencing as though automatic consequences follow from the presence or absence of particular factual circumstances. In every case, what is called for is the making of a discretionary decision in the light of the circumstances of the individual case, and in the light of the purposes to be served by the sentencing exercise.” (68)
                176 His Honour went on to say that the existence of a causal relationship between a mental disorder and the commission of an offence does not automatically produce the result that the offender will receive a lesser sentence (at 71). Nor, in my opinion, should the existence of a causal relationship between drug addiction and the commission of an offence.
                177 The circumstances in which motive may be a mitigating factor should, in my opinion, be confined to cases in which motive impinges upon the moral culpability of the offender. This can include mental, emotional or medical problems or impulsive conduct. In any event the degree to which motive may be pertinent must depend on the whole of the circumstances. (White NSWCA 23 June 1998 at 24).
                178 As noted above, it has long been the position taken in this State that addiction is not, of itself, a mitigating circumstance. This is also the position in England, (Brewster (1998) 1 CrAppR 220 at 226) Canada (Johnas (1982) 2 CCC (3d) 490 at 497). It was submitted for the Respondents to the Crown appeals that authorities in other Australian states indicate a different approach.
                179 Reliance was placed on Nolan (1998) VICCA 135 (2 December 1998). This was a majority decision to allow a severity appal. One member of the majority, Buchanan JA, referred to drug addiction in terms of “mitigating circumstances” (par 15). The other member of the majority, Butt JA, did not adopt that terminology, but rather that of “explanation” (par 19).
                180 Buchanan JA referred without disapproval to Halewyn (1984) 12 ACrimR 202 at 203 where Young CJ said:
                “… it has been said a number of times in this Court that the fact that an armed robbery is committed to support a drug habit is of little consequence in mitigating the sentence to be imposed.”
                181 Buchanan JA also referred to Bouchard (1996) 84 ACrimR 499 where at 501-502, Callaway JA quoted Hallwyn but identified a case of “a man crazed by a drug craving who committed an offence on the spur of the moment and later repented”.
                182 Nolan was such a case. It should be understood as an authority on the issue of planning/impulsiveness, and that the impulsiveness was the “mitigating circumstance”, rather than the drug addiction itself.
                183 I do not understand there to be any difference between the position in New South Wales and that in Victoria in this regard.
                184 In Douglas (1995) 56 FCR 465 the Full Court of the Federal Court heard an appeal from the Supreme Court for the Australian Capital Territory. In a joint judgment, Von Doussa, Higgins and R D Nicholson JJ said (at 470):
                “It is, of course, not a mitigating factor that a person commits a crime to feed a drug addiction. However, that is not a principle of universal application. As Jenkinson J noted, at 105, in Talbot (1992) 34 FCR 100, the moral and legal fault for the acquisition of the addiction has to be considered:
                    ‘… evaluation of moral culpability remains in my opinion as fundamental to one system of punishment for crime as evaluation of the degree of harm caused or threatened by the criminal act or actor.’
                        The age of the offender when he or she becomes addicted and the degree of judgment open to them at that age is thus relevant in evaluating the extent to which they should be punished for consequential criminal conduct.”
                185 Douglas involved an addiction that commenced at age eight and was total by age 13. This is an extreme case of little assistance in establishing a general principle that addiction lessens moral culpability. On the contrary, it affirms that self induced addiction at an age of rational choice establishes moral culpability for the predictable consequences of that choice. That was the very distinction made by the Full Court in Talbot (1992) 34 FCR 100 at 105-106.
                186 In that case, the Court took into account an addiction that was caused by medical treatment “to which his consent was at best merely formal” (105). (See also the Victorian decision in Redenbach (1990) 52 ACrimR 95 at 99). The Court referred with approval to, whilst distinguishing, the reasons of King CJ in Spiero (1979) 22 SASR 543:
                “One feels sympathy for the person who has become entangled in drug addiction, but the courts cannot treat addiction as an excuse, or even a mitigating factor, in relation to serious crime. Those who are addicted to drugs must understand that if they allow their addiction to lead them into serious crime, they must expect to receive the same serious punishment as would be received by others.” (549)
                187 In Terizakis (1986) 41 SASR 252 at 256, the Full Court reiterated this reasoning. O’Loughlin J added:
                “Despite the cravings caused by the addiction, the majority of drug addicted persons still retain, in varying measures, the ability to distinguish between right and wrong and the ability to comprehend that the consequence of armed robbery will be a lengthy term of imprisonment. To give way, out of sympathy and concern, by imposing lesser terms of imprisonment that would otherwise be imposed, might appear to be an act of humanity, but it will do nothing to stamp out the drug trade and worse, it might even give some indirect encouragement to would-be offenders.” (256)
                188 Another authority on which the Respondents to the Crown appeals rely is the decision of the Queensland Court of Appeal in Hammond (1996) 92 ACrimR 450.
                189 In Rosenberger (1996) 76 ACrimR, that Court had indicated that intoxication, whether by alcohol or drugs, would not mitigate penalty save in the case where the original addiction did not involve a free choice, (as in Talbot or Redenbach supra, the latter being quoted in Rosenberger).
                190 In Hammond the Court distinguished Rosenberger and said:
                “The true relevance of drug addiction as a factor contributing to the commission of crime and its effect in the sentencing process has never been adequately explained. In our view it is a factor that may help an offender to the extent of showing that his or her descent into the crime in question was a secondary consequence of desperation produced by a human weakness rather than a primary choice. It may be inferred in many cases (and the present case is very much in point) that the offender would almost certainly never have become a thief but for loss of control of an ordered life through drug addiction. The proposition that the results of drug addiction are self-inflicted is half true and therefore dangerous. The offender must of course be held responsible for his or her own actions. The drug addiction is not an excuse; but it is a factor that may tell the court that the real weakness of character is that of a drug addict rather than that of a robber. That may be by no means inconsequential. It is however a two-edged factor; it may also tell the court that rehabilitation is going to be difficult.
                    Just how these factors are to be applied to particular cases had not been spelled out, and it may be undesirable to attempt to do so. In our view the fact that drug addiction has led to criminal activity can properly lead to a wider understanding of the offender’s character and motivation for the crime. Depending on the circumstances of the particular case this may sometimes show the offender to be less deserving of condemnation than the primary facts, without more, might suggest.” (455-456)
                191 In its conclusion the Court returned to this theme:
                “Addiction to drugs at the time of the offence is not an excuse but may tell the court that the primary weakness of character is that of an addict rather than that of a robber. Such a factor tends to be mentioned along with an unfortunate family life as an explanation of the offender’s descent into bad habits. When the original problem was not something for which the offender was responsible, it is natural to feel some sympathy for the offender, though this must not obscure the offender’s direct responsibility for the ultimate criminal act he or she chose to do.” (467)
                192 This latter passage appears in the context of the Court listing aggravating and mitigating factors. It is noticeable that addiction is not identified as a mitigating factor. In neither passage is addiction, of itself, treated as a basis for imposing a lower sentence than would otherwise be imposed. In both passages it is emphasised that addiction is “not an excuse”. In both passages the offender’s responsibility for his or her conduct is mentioned.
                193 In my opinion, Hammond affirms that drug addiction is a relevant circumstance but is not, of itself, a mitigating factor. If, contrary to this conclusion, the remarks in Hamilton were intended to suggest otherwise - and some of the language chosen may be so interpreted - then they should be regarded as anomalous and should not be followed.
                194 The authorities are against the proposition that drug addiction should, of itself, be accepted as a mitigating factor. There is authority that where the original addiction was not a willed act, that may be taken into account by way of mitigation. The authority does not go beyond that.
                195 It is in the context of this line of authority that the submissions made for Respondents to the Crown appeals to the effect drug addiction, at least where it can be shown to be causally related to the commission of an offence, should now be accepted to be a mitigating circumstance.
                196 It was submitted that the degree of moral culpability of a particular offender is diminished by addiction. Evidence was put before the Court that there is, at least in some cases, a genetic predisposition to addiction and that addiction generally is not simply a state of mind but has a neurobiological or physiological base. It was put that an addict’s decision to perform a criminal act was not “a completely free choice”.
                197 In my opinion drug addicts who commit crime should not be added to the list of victims. Their degree of moral culpability will vary, just as it varies for individuals who are not affected by addiction. There are a number of aspects of the relationship between drug addiction and crime which indicate that moral choices are made.
                198 First is the original decision to experiment with drugs which, in the usual case, is a completely free choice. The addictive quality of drugs, together with the anti-social behaviour which so commonly results from addiction, is so widely known that persons who choose a course of addiction must be treated as choosing its consequences.
                199 Secondly, the submissions in this Court were in error in identifying the relevant conduct as the craving associated with withdrawal. The material presented to the Court did not suggest that the choice faced by addicts was between this negative feeling and the need for money to allay it. Rather the choice may often be the desire for the positive feeling said to be associated with a drug induced euphoria. The desire to bring about that state of “well-being” is, relevantly, a moral choice.
                200 Thirdly, nothing in either the process of addiction or its neuro-biological and physiological basis, leads ineluctably to the commission of crime, let alone the commission of crimes of violence against persons, such as armed robbery. Not all persons who suffer from addiction behave in this way. Those that do so, make a choice.
                201 Finally, individuals do emerge from addiction. They do so with difficulty and generally need significant amounts of help. The decision to persist with an addiction, rather than to seek assistance, is also a choice.
                202 There is no warrant, in my opinion, to assess a crime induced by a need for funds to feed a drug addiction, as being lower in the scale of moral culpability than other perceived requirements for money.
                203 Counsel making these submissions was driven to accept the proposition that an addiction to gambling, with its attendant need for money, could similarly be the basis for a claim for mitigation. He went further and indicated that an elevated sex drive which also had a physiological basis, could be a mitigating factor for the commission of a rape or for the conduct of a paedophile. In my opinion all these submissions should be rejected.
                204 It was also submitted that the principle of general deterrence should be given less weight in the context of offenders who commit their crimes for the purposes of assuaging a drug addiction. In this submission, reliance was placed on the often expressed doubts about the direct effect on potential offenders of increases in penalties imposed. Particular reliance was placed on the driven nature of a drug addicts, many of whom engage in high risk activity, like exchange of needles notwithstanding the risk of HIV infection. This kind of submission has been made many times before and I have quoted a few instances above. It has always been rejected. This Court should do so again.
                205 General deterrence always operates at the margin. Some people will continue to engage in criminal conduct notwithstanding the level of, or increases in the level of, the penalties they suffer. However, some people will be deterred. It is not to the point that some addicts engage in high risk activities. It would be necessary to establish that all addicts do so. Neither the submissions, nor the materials in support, suggest anything of this character.
                206 I attach particular significance to the impact that acknowledgment of drug addiction as a mitigating factor would have on drug use in the community. The sentencing practices of the courts are part of the anti-drug message, which the community as a whole has indicated that it wishes to give to actual and potential users of illegal drugs. Accepting drug addiction as a mitigating factor for the commission of crimes of violence would significantly attenuate that message. The concept that committing crimes in order to obtain monies to buy an illegal substance is in some way less deserving of punishment than the commission of the same crime for the obtaining of monies for some other, but legal, purpose is perverse.
                207 It may very well be the fact that increased possibility of detection has greater effect by way of deterrence than increased punishment. There is no warrant, however, for the Courts abandoning reliance on the latter. In any event the two propositions are related. It is only because detection, when it occurs, leads to a level of punishment, that increases in detection have their deterrent effect.
                208 It may very well be that the criminal justice system has a modest role to play in the control of drug addiction. But however modest that role may be, it must be performed in accordance with the basic structure of the criminal sentencing process. At the level of a structure deeply embedded in our society, not merely at the level of an individual’s calculus of risks and benefits for specific conduct, the criminal justice system is now, and has always been, based on the proposition that punishment deters and, within limits of tolerance, increased punishment has a corresponding effect by way of deterrence. This Court should not change such a longstanding assumption. Legislation would be required to alter the common law in this way.
                209 I reiterate that the process of imposing penalties for the commission of crimes, has its primary deterrent effect through its operation as a structural phenomenon of the criminal justice system. That is not capable of being assessed from the perspective of what particular penalties, or increases in penalty, may have in the case of individuals.
                210 In any event, the reasons for the guideline propounded in this judgment do not relate merely to an increase in the size of penalty. The guideline is particularly directed to overcoming the very significant proportion of cases in which non custodial sentences have been imposed. Henceforth, such sentences should be restricted to the exceptional cases to which the authorities have always referred.
                211 The deterrent effect of a sharp reduction in the proportion of non-custodial sentences which, as a result of this judgment, become custodial sentences, may well be much more significant than the deterrent effect from an increase in the level of custodial sentences. That is not to say that the latter does not have some deterrent effect but the sentencing practices, which I have analysed above, have been such as to significantly attenuate the deterrent effect of sentencing for the offence of armed robbery in the past. It is my opinion that that should change.
                212 I have now read the additional observations of Wood CJ at CL. I agree with his Honour’s judgment.
                **********


                IN THE COURT OF
                CRIMINAL APPEAL

                No. 60559/98
                60558/98
                60561/98
                60746/98
                60596/98
                60595/98
                60511/98
              SPIGELMAN CJ
              WOOD CJ at CL
              NEWMAN J
              HULME J
              SIMPSON J
                Wednesday 12 May 1999
                  REGINA v Paul Anthony HENRY
                  REGINA v Stephen Anthony BARBER
                  REGINA v Hoai Vinh TRAN
                  REGINA v Troy David SILVER
                  REGINA v Theo TSOUKATOS
                  REGINA v Bill KYROGLOU
                  REGINA v John David JENKINS
                  GUIDELINE JUDGMENT


                  213 WOOD CJ at CL: I have read in draft the reasons of Spigelman CJ. I agree with the reasons for delivering a guideline judgment in relation to the offences before the Court, and with the guidelines proposed.
                  214 I wish, however, to examine in a little more detail the submission that this Court should now declare that the fact that an offender was addicted to drugs, and committed an armed robbery in order to obtain money to support a drug habit, is a mitigating circumstance. As Spigelman CJ has observed, this would require the Court to overturn a long line of prior decisions, and to part from the position adopted in other jurisdictions, both within this country and elsewhere.
                  Drug Addiction
                  215 It cannot be gainsaid that very many offences of armed robbery are committed because of an addiction to drugs. All of the respondents to the Crown appeals presently before the Court were motivated by a drug habit. The New South Wales Bureau of Crime Statistics and Research reported in 1987 that:
                  “drug (usually heroin) dependency appears to be an important ingredient in robbery offending, and in particular in the pattern of offending of those with multiple robbery convictions.”
                  (New South Wales Bureau of Crime Statistics and Research, Robbery: Final Report (1987) at 88. When the Bureau interviewed convicted armed robbers, it found that drug use was a motivating factor in 70% of robberies (ibid at 93.)
                  216 The acceptance of the link between drug dependency and many forms of criminal activity has recently led the State of New South Wales to create a Drug Court, with a special regime for dealing with certain categories of offenders, although not those charged with offences “involving violent conduct”: Drug Court Act 1998 (NSW) S 5(2)(b).
                  217 The question whether drug addiction should operate as a mitigating factor when sentencing an offender for an offence under S 97(1) Crimes Act, or for similar offences, has been considered by this Court upon numerous occasions. The attitude taken to date is accurately reflected in Valentini (1989) 46 A Crim R 23; and in Ellis (1993) 68 A Crim R 449.
                  218 In Valentini, the Court (comprised by Maxwell, Carruthers and Loveday JJ) said, (at 25):
                  “This Court has said on countless occasions that addiction to heroin is not to be considered as a factor for the reduction of what would otherwise be an appropriate sentence for the nature of the offences which have been committed. It serves, however, to provide an explanation for the commission of the offences.”
                  219 In Ellis, Hunt CJ at CL said (at 462):
                  “The maximum sentence enacted by the legislature indicates that it expects the courts to impose severe sentences for these offences. I believe that the community has the same expectation. As was said by this Court in a bank robbery case (Petrinovic (unreported, 18 September 1990) at 9) the citizens of this community are sick and tired of armed robberies conducted by criminals in order to feed their drug habits, and this Court will support judges who recognise these as serious crimes and who impose heavy sentences for them.”
                  220 Although Kirby P (at 459) expressed his concern in Ellis at the possibility of the Court being diverted from its legal duty, by “appeals to populist notions of what the Court conceives the community to be ‘sick and tired of’”, and dissented in the decision of the majority in dismissing the appeal, that was attributable to the comparison that he thought appropriate with the sentences that had been imposed upon separate offenders in analogously similar cases. His Honour did advert to the fact that:
                  “people addicted to illegal drugs frequently have had resort to armed robberies in an attempt to secure the funds to feed the habit made expensive by its illegality”
                      but his judgment provides no encouragement for the view that such an addiction should be regarded as a factor in mitigation.
                  221 In Begnell (Court of Criminal Appeal New South Wales 26 November 1992) his Honour observed (at 4):
                  “The courts cannot, in the face of the provisions of the statutes which criminalise the conduct of the use of prohibited drugs, allow that use to become an excuse for the kind of conduct which results in the charges and convictions that bring the applicant before us today. On the other hand, it is necessary, as it seems to me, to take into account such a consideration in looking at the criminality of the action in its totality and in considering its seriousness.”
                  222 I do not understand his Honour, by the qualification contained in the second sentence of this passage, to be suggesting that the fact that an offence was committed to cater for a drug habit, should be regarded as a factor of mitigation. The contrary was, in fact, suggested by his Honour’s comments in Bradley (Court of Criminal Appeal New South Wales 26 October 1993) in a case of assault and rob involving a bag snatch from an elderly lady in a suburban street by an offender with a cocaine dependence, (at 5):
                  “… her past addiction to cocaine was not an excuse for the action of the kind which occurred on this occasion. There are organisations and individuals in the community who will help people with this form of addiction. It would be completely intolerable if people with an addiction of this kind could resort to acts of violence against elderly members of the community or indeed anyone else. The courts must make it plain that such conduct will not be tolerated.”
                  223 Similarly, in Vidler (Court of Criminal Appeal New South Wales 10 April 1986) Street CJ, with whom Lee and Lusher JJ agreed, said that:
                  “Drug addiction with the consequently generated cash need … may explain criminal activities but it cannot excuse them so as to result in diminution in sentences proper to be passed for the offences under consideration”.
                  224 In Crotty (Court of Criminal Appeal New South Wales 28 February 1994) where the offence involved was one of armed robbery of a taxi-driver, Sully J said (at p6):
                  “Drug induced crime, particularly drug induced armed robbery, will not be regarded by the Courts as being necessarily mitigated by the particular drug and alcohol problem. It needs to be said yet again that ordinary decent members of the community who look, and rightly look, to the Courts for protection of their persons and property against assault and robbery are fed up with the notion that somehow or other a person who assaults them, who steals their property, who breaks into and ransacks their homes, is somehow not to be blamed for what he or she has done because of the presence of some drug or alcohol or like sociological problem invoked to explain what has been done in a particular case. I do not suggest, of course, that the presence of a drug or alcohol problem is not a relevant consideration.
                      All I am seeking to say is that, as with the bare fact of youth, so also with the bare fact of drug and alcohol inducement, it must not be allowed to become a cloak of convenience behind which people who are in every relevant sense adults can shelter from the responsibility for their serious breaches of criminal law.”
                  225 More recently it was said, by McInerney J, with whom Bruce J agreed, in Jebara (Court of Criminal Appeal New South Wales 2 August 1995):
                  “It has been pointed out that whatever sympathy the courts may feel for a person who is subject to a compulsive and expensive habit, it cannot be made an excuse for crime. The community has to be protected against criminal action, whatever the motivation may be. In R v Martin (Court of Criminal Appeal New South Wales 19 March 1992 unreported) Hunt CJ at CL referring to such a habit providing a reason for the commission of offences, stated it provided no excuse in mitigation. At best, such a circumstance avoids a finding of aggravation by reason of the offence being committed for pure greed."
                  226 This approach, which has been referred to as a “hard line” approach (Potas, Sentencing Robbers in New South Wales 1990 at 242, 163 can also be seen in decisions such as: Atkins (Court of Criminal Appeal New South Wales 27 May 1999); Hines (Court of Criminal Appeal New South Wales 25 May1998); Salameh (Court of Criminal Appeal New South Wales 9 June 1994); Crotty (Court of Criminal Appeal New South Wales 28 February 1994 unreported); Stanford (Court of Criminal Appeal New South Wales 23 November 1993); Jansz (Court of Criminal Appeal New South Wales 30 September 1993); Shinner (Court of Criminal Appeal New South Wales 10 July, 1992); and Cottier (Court of Criminal Appeal New South Wales 18 December 1990).
                  227 A departure from this line of authority can be seen in the observation of Hulme J, in Murray (Court of Criminal Appeal New South Wales 7 November 1995) where his Honour said that:
                  “the fact that (the applicant’s second criminal) career was inspired by a need for funds to support his heroin habit, is, to my mind, a mitigating factor”.
                  228 He was, however, in the minority on this point, Grove J, with whom Hunt CJ at CL agreed, saying:
                  “I would not regard an explanation founded in a need to support a heroin habit as a mitigating factor in regard to these offences. No doubt the habit offers explanation but in my view, it does no more than that.”
                  229 A similar stand has been taken elsewhere: See, for example, Lawrence (1988) 10 Cr App 464; Brewster (1998) 1 Cr APP R 220, at 226, and Johnas (1982) 2 CCC (3d) 490, at 497; and Terizakis (1986) 41 SASR 252 at 256 and also at 252 where Zelling ACJ said:
                  “It cannot be too strongly stated that heroin addiction and the need to obtain money in furtherance of it, is not a matter in mitigation in charges of robbery under arms.”
                  230 Reference may also be made to Spiero (1979) 22 SASR 543 at 5488-549, where King CJ with whose reasons for judgment Walters and White JJ concurred, said:
                  “one feels sympathy for a person who has become entangled in drug addiction but the Court cannot treat an addiction as an excuse, or even a mitigating factor, in relation to serious crime. Those who are addicted to drugs must understand that if they allow their addiction to lead them into serious crime, they must expect to receive the same severe punishment as would be received by others.”
                  231 In Halewyn (1984) 12 A Crim R 202, Young CJ, in whose judgment Kaye and Beach JJ concurred, said (at 203):
                  “The offence thus committed was a very serious one which, as I have said, is all too prevalent in this community. It is an offence which the courts are bound to deal with by imposing substantial sentences, and it has been said a number of times in this Court that the fact that an armed robbery is committed to support a drug habit is of little consequence in mitigating the sentence to be imposed.”
                  232 In Talbot (1992) 34 FCR 100, Jenkinson J, with whose reasons O’Loughlin and Higgins JJ agreed, said at 105-106:
                  “What in my opinion justifies the denial to the addict of leniency in recognition of the relative strength of his temptation to obtain his drug by armed robbery are on the one hand the magnitude of the community’s need of protection from that offence and on the other hand the moral and legal fault which originated the addiction. It seems unfashionable - and, having regard to recently acquired knowledge of molecular biology, perhaps imprudent - to express moral judgments in sentencing criminals. But evaluation of moral culpability remains in my opinion as fundamental to our system of punishment for crime as evaluation of the degree of harm caused or threatened by the criminal act or actor. So much I think the reasons of Mason CJ, Brennan, Dawson and Toohey JJ for their judgment in Veen v The Queen (No 2) (1988) CLR 465 make clear, particularly in their quotation (at 473-474) from an article by C S Lewis (Even more worthy of judicial citation is Lewis’ The Abolition of Man (2nd ed 1946).”
                  233 Their Honours did, however, in this case note the special circumstance where the addiction of the offender originated in the administration of a potentially addictive substance in the course of medical treatment, to which his or her consent was, at best, merely formal. The moral culpability of such an offender was said to be different from that of the offender whose addiction originated in the voluntary administering of a substance that he knew (or I would add, should have known) to be addictive, and the use of which is proscribed by the criminal law.
                  234 In Hammond (1997) 2 Qd R 195, the Court said at 199-200:
                  “The true relevance of drug addiction as a factor contributing to the commission of crime and its effect in the sentencing process has never been adequately explained. In our view it is a factor that may help an offender to the extent of showing that his or her descent into the crime in question was a secondary consequence of desperation produced by a human weakness rather than a primary choice. It may be inferred in many cases (and the present case is very much in point) that the offender would almost certainly never have become a thief but for loss of control of an ordered life through drug addiction. The proposition that the results of drug addiction are self inflicted is half true and therefore dangerous. The offender must of course be held responsible for his or her own actions. The drug addiction is not an excuse; but it is a factor that may tell the court that the real weakness of character is that of a drug addict rather than that of a robber. That may be by no means inconsequential. It is however a two edged factor; it may also tell the court that rehabilitation is going to be difficult.
                      Just how these factors are to be applied to particular cases has not been spelled out, and it may be undesirable to attempt to do so. In our view the fact that drug addiction has led to criminal activity can properly lead to a wider understanding of the offender’s character and motivation for the crime. Depending on the circumstances of the particular case this may sometimes show the offender to be less deserving of condemnation than the primary facts, without more, might suggest.”
                  235 Later, the Court said at 211:
                  “Addiction to drugs at the time of the offence is not an excuse, but may tell the court that the primary weakness of character is that of an addict rather than that of a robber. Such a factor tends to be mentioned along with an unfortunate family life as an explanation of the offender’s descent into bad habits. When the original problem was not something for which the offender was responsible, it is natural to feel some sympathy for the offender though this must not obscure the offender’s direct responsibility for the ultimate criminal act he or she chose to do.”
                  236 These observations were relied upon by the respondents as providing some support for the proposition that the fact of drug addiction may properly lead to the culpability of the offender being regarded as less deserving of condemnation than would otherwise be the case. However, it is clear that the Court in Hammond did not question the widely accepted principle that drug addiction is not an excuse, and emphasised that the relevance of drug dependence will depend upon the circumstances of the particular case. The decision does not stand as authority for the general proposition advanced.
                  237 Nolan (1998) VSCA 135 also provides only limited assistance to the respondents. Although Buchanan JA referred to drug addiction as a “mitigating” circumstance, the other member of the majority, Butt JA spoke of it as an “explanation”. It was an extreme case, where the impulsiveness of the offence, rather than the mere fact of drug addiction, took the matter outside the general rule recognised in Halewyn.
                  238 Douglas (1995) 56 FCR 465 similarly was an extreme case. Von Doussa, Higgins and Nicholson JJ there accepted that, while it is not a mitigating factor that a person commits a crime to feed a drug addiction, nevertheless there will be extreme cases where that principle will not apply. The moral and legal fault for the acquisition of the addiction can properly be considered, for example where the offender becomes addicted at an age when the “degree of judgment” open to him or her is limited.
                  239 The submission advanced essentially turned upon the proposition that, within the context of the national concern that currently exists in relation to the scourge of drug dependency, and its effects not only upon individual users and their families, but also upon the community as a whole, some greater flexibility should be exercised in sentencing those whose criminality is substantially linked to an addiction.
                  240 In support of this submission reference was made to the views expressed by some academics and researchers that imprisonment provides no deterrent for this class of offender, and may tend only to exacerbate the problem.
                  241 Additionally, reference was made to the need to look for new solutions or creative responses for what has proved to be an intractable pattern. Illustrations cited as either having been adopted or under review in this area include the Drug Court, expanded access to rehabilitation programmes, heroin trials, safe injecting rooms, and new forms of chemical therapy in the place of, or as an alternative to methadone (naltrexone, acamprosate, buprenorphine, slow release oral morphine and so on.)
                  242 Within that context it was submitted that the bare statement that drug addiction is “not an excuse,” or is “not a factor in mitigation” reflects an approach by the courts which is too rigid. On the contrary, it was submitted a more constructive approach should now be adopted which gives greater recognition to the cravings of drug dependent offender who is forced through the absence of suitable alternatives, to resort to crime to feed a habit.
                  243 Reference to the observation of Mildren J in Lewfatt (1993) 66 A Crim R 451, concerning the imprisonment of the typical offender with which the guideline judgment deals, underlies, at least in part this submission. His Honour there said:
                  “The prisoner was young, and had undergone significant changes in her outlook. Her prior convictions were drug and alcohol related. She was now determined to stay off drugs and alcohol, and to pursue her ambitions for a career, and accept responsibility as a parent. Her history strongly suggested that she had gone through a difficult period of immaturity, and was now ready to rehabilitate herself. In those circumstances a period of actual imprisonment could well be counterproductive to this process. If one of the main purposes of punishment is to protect society, society’s interests are best served by a sentencing disposition which promotes the rehabilitation of the prisoner, rather than a disposition which may have the opposite effect.”
                  244 To this statement can be added the observation of the majority of the High Court in Veen No.2 (1988) 164 CLR 465 at 476:
                  The purposes of criminal punishment are various: protection of society, deterrence of the offender and of others who might be tempted to offend, retribution and reform. The purposes overlap and none of them can be considered in isolation from the others when determining what is an appropriate sentence in a particular case. They are guideposts to the appropriate sentence but sometimes they point in different directions.”
                  245 In advancing the proposition that drug dependence should be recognised as a significant factor of mitigation, justifying a lesser sentence than might otherwise be appropriate, the respondents sought to draw upon material questioning the extent to which such dependence is a matter of full and free choice. An affidavit was tendered, for this purpose, sworn by Dr. Jurd, a psychiatrist specialising in the treatment of drug offenders, in which he deposed that:
                  “drug addiction has a well documented neurobiological or physiological basis.”
                  246 In addition he said that the:
                  “predisposition to addiction has a substantial physiological (genetic) component. Thirty years of study of the neurobiological basis of alcoholism has revealed time and again that having a positive family history is among the most powerful predeterminants of alcoholism:
                      (a) identical twins are more likely than fraternal twins to both suffer from alcoholism.
                      (b) adopted away children of alcoholic parents, even if they have never lived with their parents, are more likely to themselves become alcoholic than other adoptees.
                      (c) the risk of developing alcoholism increases depending on how many alcoholic relations one has, whether or not you live with them.
                      (d) a series of molecular biological experiments have shown that at least one gene is over-represented among alcohol dependent people.
                      There is every reason to extrapolate this information to other drugs. Similar studies have been done with drug dependents producing very similar results.”
                  247 He continued by noting that drug addiction affects the addicts’ behaviour in predictable ways, altering values and overwhelming other priorities as the addiction becomes more severe.
                  248 The hypothesis that addiction has a genetic component or predisposition was not tested. Nor was the validity of the comparison with alcohol dependency. Whether this view be correct or not, I am not persuaded that there is an inevitable causal relationship or progression from drug addiction to the commission of criminal offences, or to an incremental increase in the seriousness of these offences.
                  249 In many instances, as studies tend to show, that may well be he case: Dobinson et al Drugs and Crime: A Survey of New South Wales Property Offenders 1984. New South Wales Bureau of Crime Statistics and Research, 56, 57, 62 and 64; Dobinson Making Sense of the Heroin-Crime link (1882) 22 ANZJ Crim, 269; Maher et al Running the Risks: Heroin, Health and Harm in South West Sydney National Drug & Alcohol Research Centre University of New South Wales Monograph No 38, 51 and 62; Hall, Methadone Maintenance Treatment as a Crime Control Measure Bureau of Crime Statistics and Research (1996) 2; Hall The Role of Legal Coercion in the Treatment of Offenders with Alcohol and Heroin Problems (1997) 30 ANZJ Crim 105; Lipton The Effectiveness of Treatment for Drug Abusers under Criminal Justice Supervision (1995) National Institute of Justice, 4.
                  250 However, many offenders commence their criminal careers before turning to drugs, and not all persons who become addicted to drugs commit armed robberies or even serious property offences.
                  251 Moreover, there are many factors other than drug dependency, such as multiple social disadvantage or an attachment to a criminal sub culture, that contribute to criminal behaviour; and it would, accordingly, be too simplistic to generalise that drug dependency is inevitably the precursor to or cause of criminality or of recidivism.
                  252 Nor am I persuaded of the appropriateness of the suggested analogy between drug addiction and mental abnormality, in respect of which the element of general deterrence is often given less weight: Veen (No.2) at 476-477, Scognamiglio (1991) 56 A Crim R 81; Tsaris (1996) 1 VR 398; Letteri (Court of Criminal Appeal New South Wales 18 March 1993) and Engert (1995) 84 A Crim R 67.
                  253 The relevant principle as stated in Letteri by Badgery-Parker J, in a passage adopted by Gleeson CJ in Engert is as follows:
                  “… that whereas general deterrence is a relevant consideration in every sentencing exercise, it is a consideration to which less weight should be given in the case of an offender suffering from a mental disorder or severe intellectual handicap. In an extreme case, the proper application of this principle may produce the result that considerations of general deterrence are totally outweighed by other factors. In every case it is a matter of balancing the relevant factors in a manner no different from that which is involved in every sentencing exercise.”
                  254 The reason for this approach lies in the circumstance that the community will readily understand that the offender who suffers from a mental disorder or abnormality is less in control of his or her cognitive facilities or emotional restraints, and in some instances lacks the ability to make reasoned or ordered judgments. Almost invariably there is a limited appreciation of the wrongfulness of the act, or of its moral culpability, which although falling short of avoiding criminal responsibility does justify special consideration upon sentencing. Moreover, such a condition is inherent and its presence does not depend upon any element of choice.
                  255 The case of the drug dependent offender is very different in so far as the community is likely to see such a person as a perfectly appropriate medium to stand as an example to others.
                  256 These differences were implicitly recognised in Terizakis (1986) 41 SASR 252 where O’Loughlin J, with whose reasons Cox J agreed, said at 256:
                  “… heroin addiction cannot, of itself, justify a sentencing judge reducing a penalty, which is otherwise appropriate to the offence that has been committed and to the circumstances in which the offence was committed. One cannot help but have immense sympathy for a person who is addicted to heroin; and it would be callous to adopt an attitude that, because it is self-inflicted, the addict is not worthy of sympathy or concern. But having said that, it unfortunately remains a fact of life that all too many armed hold-ups are these days committed by persons who, in one way or another, are dependent upon drugs and are seeking money to meet the costs of that dependency. Despite the cravings caused by the addition and despite irrationality that is so often a consequence of the addiction, the majority of drug addicted persons still retain, in varying measures, the ability to distinguish between right and wrong and the ability to comprehend that the consequence of armed robbery will be a lengthy term of imprisonment. To give way, out of sympathy and concern, by imposing lesser terms of imprisonment than would otherwise be imposed, might appear to be an act of humanity, but it will do nothing to stamp out the drug trade and, worse, it might even give some indirect encouragement to would-be offenders.”
                  257 The comparison is also imperfect in so far as it overlooks the original element of choice every person has, initially whether or not to experiment with drugs, and thereafter to continue with their use. Even if some persons have a genetic predisposition or vulnerability to addiction, they are not, in my view, entitled to claim any favours in that regard. The addictive quality of drugs and the potential of a slide into a degraded or criminal lifestyle are so well known as to invite a free choice. Those who choose to use drugs must, in my view, accept the consequences.
                  258 Moreover, as Gleeson CJ observed in at 71:
                  “…the question of the relationship, if any, between the mental disorder and the commission of the offence, goes to circumstances of the individual case to be taken into account in the application of the relevant principles. The existence of such a causal relationship in a particular case does not automatically produce the result that the offender will receive a lesser sentence, any more than the absence of such a causal connection produces the automatic result that an offender will not receive a lesser sentence in a particular case. For example, the existence of a causal connection between the mental disorder and the offence might reduce the importance of general deterrence, and increase the importance of particular deterrence or of the need to protect the public. By the same token, there may be a case in which there is an absence of connection between the mental disorder and the commission of the offence for which a person is being sentenced, but the mental disorder may be very important to considerations of rehabilitation, or the need for treatment outside the prison system.”
                  259 By the same process of reasoning the existence of a causal relationship between a drug habit and the commission of an offence should not, as a matter of general principle, automatically lead to a lesser offence.
                  260 The respondents also relied upon the proposition that there was little evidence that the punishment of offenders has any general deterrent effect, whether for offenders at large or drug dependent persons in particular. It may be recognised that while there are different views about this, some studies have been interpreted as showing that the perceived severity of a sentence, as distinct from the certainty of detection and arrest, does not of itself provide a deterrent effect: Anderson et al “Formal and Informal Sanctions: A comparison of deterrent effects” (1997) Social Problems 25: 103-117; Paternoster et al “Perceived Risk and Social Control: Do Sanctions really deter?” (1983) Law and Society Review 17: 467-479; Baron et al “Deterrence and Homeless Street Youths (1998) Canadian Journal of Criminology 27-60; James Austin et al “Does Imprisonment Reduce Crime? A Critique of ‘Voodoo’ Criminology” (1993) National Council on Crime and Delinquency. 10, 22.
                  261 In the case of drug dependent persons contemplating the commission of an armed robbery, in particular, it was suggested, that the prospect of imprisonment provides little deterrent, in that:
                  · their addiction causes them to give high priority to drug seeking behaviour, sometimes to the exclusion of all else;
                  · they are accustomed to engaging in high risk activity: Maher et al “Running the Risks: Heroin, Health and Harm in South West Sydney” National Drug and Alcohol Research Centre University of NSW NDARC Monograph no. 38 at 26, l20, 125; Hall Methadone Maintenance Treatment as a Crime Control Measure (1996) Bureau of Crime Statistics and Research, 2.
                  · the clearance rate for armed robbery is less than 1 in 5 (Chilvers: New South Wales Recorded Crime Statistics 1998 at 30-32;
                  · there is no realistic prospect of law enforcement activity reducing the size of the drug addicted population or of it reducing the street level price of drugs (thereby alleviating the financial pressure to resort to crime to support a habit) Maher supra at 126) and
                  · despite the steady increases in sentences, following the introduction of the Sentencing Act 1989, and also amendment of the Crimes Act 1900 to increase the maximum available penalty, in respect of many offences, the crime rate has not fallen.
                  262 It was additionally submitted, there was little evidence that lengthy periods of imprisonment have any substantial rehabilitative consequences for the drug dependent offender. To the contrary, the argument ran, and as recidivism rates suggest, incarceration may be counterproductive, initiating or encouraging the offender into criminal behaviour, and breaking social and familial bonds which militate against such behaviour.
                  263 Alternative sentencing options, it was submitted, were more likely to reduce drug dependency and the consequent incidence of criminality and of recidivism: Hall: “Methadone Maintenance Treatment as a Crime Control Measure” (June 1996) Contemporary Issues in Crime and Justice, Bureau of Crime Statistics and Research 3, and 6-7; Hall: “The Role of Legal Coercion in the Treatment of Offenders with Alcohol and Heroin Problems” (1997) Vol 30 ANZJ Crim 113; Lipton: “The Effectiveness of Treatment for Drug Abusers Under Criminal Justice Supervision” (1995) National Institute of Justice 51-53; Murphy: “Drug Courts: An Effective Efficient Weapon in the War on Drugs” (1997) Illinois Bar Journal Vol 85 487; Tauber: Drug Courts: a Judicial Manual California Centre for Judicial Education and Research (1994) 1-2 and 9-10; and Belenko “Research on Drug Courts: A Critical Review” (1998) National Drug Court Institute Review Vol 1 Issue 1, 21-23, 29 and 35.
                  264 The views expressed in these studies concerning the deterrent value of imprisonment, and the possible exposure of prison inmates to influences that may increase their prospects of recidivism, are not confined to drug dependent offenders. They are of common application, particularly in the case of first offenders, for whom the slamming prison door principle remains apposite, and also in the case of those who are yet to offend, for whom the prospect of imprisonment has a particular relevance.
                  265 I am not prepared to advocate any departure from the long accepted wisdom that imprisonment does have a personal and general deterrent effect. It is a notion deeply entrenched in the criminal law, and it has the imprimatur of the legislature which has prescribed significant maximum penalties for the offence under consideration, as well as for other offences involving serious criminality.
                  266 Moreover, it cannot necessarily be assumed from the fact that increases in sentences have not been accompanied by any noticeable drop in crime rates, that they lack deterrent effect. In the absence of any control, it cannot be known whether that crime rates would have been higher had sentences not been increased.
                  267 The risk of exposure to undesirable influences within a corrective environment, similarly cannot be used as a justification for abandoning, or even limiting imprisonment, as a general response to criminal conduct. There remains for every offender a choice between reform and recidivism, and the problem is better addressed by the development of adequate programs and rehabilitation options within the prison environment, than it is by a significant change in sentencing policy.
                  268 While it may be accepted that the rehabilitative aspect of sentencing is not an unimportant consideration: Osenkowski (1982) 30 SASR 21; Douglas (Court of Criminal Appeal New South Wales 4 March 1997); Eastway (Court of Criminal Appeal New South Wales 19 May 1992); Fabian (1993) 64 A Crim R 365; and Halewyn, I am not persuaded that the offender, whose criminality was directly influenced by drug addiction, should be placed into a special category for sentencing purposes.
                  269 The objective of rehabilitation is but one aspect of sentencing. It needs to be taken into account along with the objectives of retribution and deterrence.
                  270 It is more appropriate, in my view, to have regard to these considerations, upon an individual case basis, in which it is recognised that the offence of armed robbery remains a very serious crime, which, save in exceptional circumstances, calls for full time imprisonment. That is not to say that, in an individual case, the fact of drug dependency of the person standing for sentence may not remain a relevant fact reflecting on the objective circumstances of the offence, and/or the subjective circumstances of the offender, and particularly on the relativity of the minimum and additional terms: Bugmy (1990) 169 CLR 525 at 537; and B (1993) 68 A Crim R 547.
                  271 This approach accords with the caution offered by Gleeson CJ in Engert at 68:
                  “It is … erroneous in principle to approach the law of sentencing as though automatic consequences follow from the presence or absence of particular factual circumstances. In every case, what is called for is the making of a discretionary decision in the light of the circumstances of the individual case, and in the light of the purposes to be served by the sentencing exercise.”
                  272 It also avoids the problems associated with establishing categories according to the offender’s need for money (e.g. to feed a drug addiction, to support a gambling habit, to pay off business or family debts, to establish a business, etc) which would lend themselves to abuse and to moral judgments in respect of which minds may legitimately differ.
                  273 In my view the relevant principles are as follows:
                      (a) the need to acquire funds to support a drug habit, even a severe habit, is not an excuse to commit an armed robbery or any similar offence, and of itself is not a matter of mitigation;
                      (b) however the fact that an offence is motivated by such a need may be taken into account as a factor relevant to the objective criminality of the offence in so far as it may throw light on matters such as:
                          (i) the impulsivity of the offence and the extent of any planning for it; (cf Bouchard (1996) 84 A Crim R 499 at 501-2); and Nolan (1988) VSCA 135 (2 December 1998);
                          (ii) the existence or non existence of any alternative reason that may have operated in aggravation of the offence, eg that it was motivated to fund some other serious criminal venture or to support a campaign of terrorism;
                          (iii) the state of mind or capacity of the offender to exercise judgment, eg if he or she was in the grips of an extreme state of withdrawal of the kind that may have led to a frank disorder of thought processes or to the act being other than a willed act;
                      (c) It may also be relevant as a subjective circumstance, in so far as the origin or extent of the addiction, and any attempts to overcome it, might:
                      (i) impact upon the prospects of recidivism/rehabilitation, in which respect it may on occasions prove to be a two-edged sword (eg Lewis Court of Criminal Appeal New South Wales 1 July 1992);
                      (ii) suggest that the addiction was not a matter of personal choice but was attributable to some other event for which the offender was not primarily responsible, for example where it arose as the result of the medical prescription of potentially addictive drugs following injury, illness, or surgery (cf Hodge Court of Criminal Appeal New South Wales 2 November 1993; and Talbot); or where it occurred at a very young age, or in a person whose mental or intellectual capacity was impaired, so that their ability to exercise appropriate judgment or choice was incomplete;
                      (iii) justify special consideration in the case of offenders judged to be at the “cross roads”: Osenkowski(19882) 5 A Crim R 394.
                  274 To go further, and to accept the fact of drug addiction as a mitigating factor generally, would not be justified in principle. Moreover, it would involve an exercise in irresponsibility on the part of the Court, if it were understood as a message that committing the crime of armed robbery to feed a drug habit is less deserving of censure than would otherwise be the case.
                  275 The legislature has, by the heavy maximum penalty prescribed for armed robbery, spoken clearly in relation to this offence. Drug dependent persons should not be encouraged, as a class, to think that they are free to engage in serious criminal conduct of whatever kind with impunity, or with any hope of favourable treatment because they are able to show that they needed money through their addiction.
                  276 In summary, I see no reason to depart from the planks of punishment, retribution and rehabilitation that underlie the sentencing process generally, and that permit of individualised sentencing by reference to the objective and subjective circumstances of each case.
                  277 I do not suggest by these observations that greater attention and resources should not be dedicated to drug rehabilitation generally, both within and outside the prison environment. Clearly it is desirable that alternative approaches to the problem be trialed, it being far preferable that potential criminality be headed off than punished after the event.
                  **********

                  IN THE COURT OF

                  CRIMINAL APPEAL
                                      60559/98
                                      60558/98
                                      60561/98
                                      60746/98
                                      60596/98
                                      60595/98
                                      60511/98
                                                  SPIGELMAN CJ
                                                  WOOD CJ at CL
                                                  NEWMAN J
                                                  HULME J
                                                  SIMPSON J
                                              Wednesday 12 May 1999

                            REGINA v Paul Anthony HENRY
                            REGINA v Stephen Anthony BARBER
                            REGINA v Hoai Vinh TRAN
                            REGINA v Troy David SILVER
                            REGINA v Theo TSOUKATOS
                            REGINA v Bill KYROGLOU
                            REGINA v John David Jenkins

                    JUDGMENT

                    278 NEWMAN J: I have read the reasons of Spigelman CJ relating to the implementation of guide lines in respect of sentencing in relation to the crime of armed robbery and whether drug addiction can be taken into account by way of mitigation. I agree with those reasons. I also agree with Wood CJ at CL.
                    **********

                    IN THE COURT OF
                    CRIMINAL APPEAL
                                        60559/98
                                        60558/98
                                        60561/98
                                        60746/98
                                        60596/98
                                        60595/98
                                        60511/98
                      SPIGELMAN CJ
                                                  WOOD CJ AT CL
                                                  NEWMAN J
                                                  HULME J
                      SIMPSON J
                      Wednesday 12 May 1999
                      REGINA v Paul Anthony HENRY
                      REGINA v Stephen Anthony BARBER
                      REGINA v Hoai Vinh TRAN
                      REGINA v Troy David SILVER
                      REGINA v Theo TSOUKATOS
                      REGINA v Bill KYROGLOU
                      REGINA v John David Jenkins

                      GUIDELINE JUDGMENT

                      HULME J:
                      Guideline Judgment
                      279 I have read in draft the Reasons of Spigelman CJ. I agree with his Honour’s conclusion that a guideline judgement is required and generally with His Honour’s reasons for that conclusion. However, on that topic there are some additional remarks I would make.
                      280 Firstly, I am satisfied that the incidence of armed robbery has had an impact on society by changing patterns of behaviour out of a sense of apprehension. The increasing use over recent, perhaps 10 or so, years of time delay locks, video cameras and of the practice that safes can only be opened by employees of companies such as Brambles with their armoured vans, I have no doubt, is a reaction to this type of conduct.
                      281 Secondly, the evidence adduced by the Crown satisfies me that there has been a substantial increase in the incidence of armed robbery. In this regard four sources of data were presented to the court: Victim Surveys, Court Statistics, Statistics kept by the NSW Bureau of Crime Statistics and Research, and information in the Report on Government Services 1999 released by the Steering Committee for the Review of Commonwealth/State Service Provision.
                      Victim Survey Data
                      282 The Victim Survey data was contained in annexure B14 to the Crown’s original submissions. This was a May 1998 “Crime and Justice” bulletin published by the NSW Bureau of Crime Statistics and Research. The article in that bulletin deals with “robbery” generically and records that it is based on what was said to be “a representative sample of about 6,500 NSW households involving some 13,000 individuals”. It notes however that there was some “confusion about the definition of robbery (resulting) in a significant over estimate of the incidence of robbery in NSW between 1990 and 1995”. A change in the definition of robbery was then used in the survey with the result that data collected in 1996 and 1997 is not comparable with that collected earlier.
                      283 The article purports to record victimisation rates for robbery between 1990 and 1997. The figures for 1996 and 1997 are recorded as 0.4% and 0.3% respectively. The figures for other years have to be derived from a rather small chart and, although precision is impossible, seem to be as set out below. The direction of movement from year to year is a correct reflection of the chart.
                      1990 1.7
                          1991 1.6
                          1992 1.1
                          1993 1.3
                          1994 1.2
                          1995 1.8
                      284 The article also recorded that in 1997 over 65% of the respondents to the survey who had been a victim of a robbery indicated that there was no weapon used and less than 3% of the robbery victims indicated that a firearm had been used.
                      Criminal Court Statistics
                      285 These were extracted from the relevant pages of seven annual publications produced by the NSW Bureau of Crime Statistics and Research and were:-


                      YearNumber of Persons ChargedNumber of Charges Finalised
                      19918531306
                      19929701481
                      19938341251
                      19947681051
                      1995681943
                      1996708976
                      19977841128
                      286
                      In its submissions the Crown said that “the publication records the number of people charged separated by the principal offence committed by each person (i.e. the most serious offence) and the total number of charges finalised”. However, reference to page 54 of the full 1996 publication suggests that this is not so and that, whatever other counts a person may have faced, so long as one count was robbery that person is included within the “number of persons charged”. I have not found a clear explanation of the “number of charges finalised”. It is not apparent whether it reflects the number of counts of robbery dealt with or the number of counts dealt with in the case of offenders who faced one count of robbery.
                      NSW Bureau of Crime Statistics and Research
                      287 The third set of statistics are from publications of the NSW Bureau of Crime Statistics and Research recording information supplied by the police. The Court was informed that methods of recording and correlation have changed over the years and that it is not possible to directly compare figures prepared under one system with those prepared under another. Although one cannot directly compare the figures in one of the periods referred to below with the figures in another, it is possible to compare figures within a period with other figures in the same period.
                      1982/83 - 1989/1990
                      288 Annexure B11 to the Crown’s original submissions was an extract from the NSW Recorded Crime Statistics 1989. It shows the following rate of recorded offences per 100,000 population.
                      Offence82/8383/8484/8585/8686/8787/8888/8989/90
                      Robbery with firearm32.522.120.315.516.417.816.514.4
                      Robbery with a weapon not a firearm12.09.812.515.416.719.519.722
                      Total rows 1 & 244.631.932.830.933.137.236.236.4
                      Robbery - no weapon37.637.436.030.137.338.741.644.2
                      Total rows 3 & 482.269.368.861.070.475.977.880.6

                      Note: The total of rows 3 and 4 has been inserted by me.

                      289 So far as armed robbery is concerned, the figure for 1982/83 is so high it looks to be an aberration. The figures show an increase of about 13% between 1983/84 and 1989/90 or between the three year period 83/84 - 85/86 on the one hand and the three year period 87/88 - 89/90 on the other. The figures for “robbery - no weapon” are suggestive - I put the matter no higher - of an increasing trend.
                      1991-1993
                      290 The information for 1991/93 is much briefer. It comes from the Bureau publication NSW Recorded Crime Statistics 1993 and is contained in Annexures B9-B10 to the Crown’s original submissions. The commentary records that for robbery overall “there was no statistically significant upward or downward trend over the 3 year period to December 1993”. However, it is recorded that “there was a statistically significant downward trend in robbery with a firearm … a 30.7% decrease”.
                      April 1994-December 1995
                      291 The information covering the period April 1994 to December 1995 is again brief. The Court was informed that it came from an August 1996 Crime and Justice Bulletin published by the Bureau. The extract forms Annexure B8 to the Crown’s original submissions. The commentary records:-
                      “From the first 9 months to the last 9 months of the period April 1994 to December 1995, robbery without a weapon increased by 7.3% and robbery with a weapon other than a firearm went up by 27.3%. The recent trend for robbery with a firearm has been neither upward or downward.”
                      292 It is not possible from the evidence to say whether this information relates to the absolute number of offences or is related to population. However, one can be confident that the population of New South Wales did not increase at the rate of 27.3% in this period.
                      1995-1998
                      293 Information for 1995-97 comes from the Bureau publication NSW Recorded Crime Statistics 1997. It is contained inter alia in Annexures B4-B6 of the Crown’s original submissions. This is extended to 1998 in an extract from the 1998 publication which formed Annexure C to the further Crown submissions. Apart from a decrease between 1997 to 1998 in the case of robbery with a firearm and an (insignificant) decrease in the case of robbery without a weapon in the same period, these figures show a substantial increase in offences. The figures per 100,000 population are:-
                      Offence1995199619971998
                      Robbery with firearm11.113.317.5/18.214.4
                      Robbery with weapon not firearm2430.553.4/53.969.2
                      Total rows 1 & 235.143.870.9/72.183.6
                      Robbery without a weapon73.979.8102.2/103.4101.9
                      Total rows 3 & 4109123.6173.1/175.5185.5

                      Notes: Totals have again been inserted by me.
                          Annexures B4-B6 on the one hand and Annexure C on the other state different figures for 1997.
                      294 At face value these figures indicate that between 1995 on the one hand and 1997 and 1998 on the other there has been an increase of over 100% in the incidence of armed robbery, more so where the “arm” has not been a firearm.
                      295 Some documents show the absolute number of offences committed during some of the periods referred to above. Annexure B12 to the original Crown submissions and Annexures B4-B6 and C to the further Crown submissions reveal:-

                      Offence82/8383/841995199619971998
                      Robbery no weapon20042009392643275633/56925511
                      Robbery with firearm17331187597732 962/1001 821
                      Robbery with weapon not firearm 641 529128916562951/29783861

                      Note: Annexures B4-B6 on the one hand and Annexure C on the other state different figures for 1997.

                      296 I do not forget the remarks made above about the difficulty of comparing figures for one period with figures in another . However, these figures show that the number of offences of robbery without a weapon has at all times been a substantial proportion of the total, and during 1995-1998 significantly exceeded the total of the number of offences involving a weapon whether or not a firearm. It follows from these figures that the Victim Survey Data and Criminal Court statistics which refer to robberies generically are of no assistance in making any judgment on the question of whether there has been a change in the incidence of armed robbery.
                      Report on Government Services 1999
                      297 This document is Annexure A to the Crown’s supplementary submissions. It records that on the basis of crimes reported to police the incidence of armed and unarmed robbery per 100,000 of population in Australia and its various states and territories during 1993 to 1997 was as follows:-

                      Armed Robbery
                      YearNSWVICQLDWASATASACTNTAUST
                      199338.622.230.328.333.59.618.18.330.2
                      199436.717.327.034.331.813.321.38.228.3
                      199562.316.826.638.724.812.925.79.836.7
                      199648.618.627.254.823.110.131.814.334.1
                      199779.325.736.658.925.39.536.212.248.6
                      Unarmed Robbery
                      YearNSWVICQLDWASATASACTNTAUST
                      199361.520.532.029.782.912.920.419.542.3
                      199484.819.033.537.371.318.021.922.850.0
                      199594.521.132.942.274.413.527.033.454.5
                      199693.623.436.542.367.420.035.145.155.3
                      1997121.027.835.059.357.422.037.426.766.1


                      298 The combined figures for New South Wales, per 100,000 population are set out in the table below. For the purposes of comparison, I have set out in a third column the comparable Bureau of Crime Statistics figures where available.
                      1993 101.1
                          1994 121.5
                          1995 156.8 109
                          1996 142.2 123.6
                          1997 200.3 173.1/175.5
                      299 Albeit there are some “dips”, the figures for New South Wales show a decided upward trend. Except for South Australia and, in the case of armed robbery, Tasmania, the other states and territories also show an upward trend in both classifications. Australia wide, including New South Wales, there was an increase in each category between 1993 and 1997 of over 50%.
                      Conclusion
                      300 The inconsistency between the above sets of figures raises the question whether any can be relied upon. At first blush each has doubt thrown on it by one or more of the others.
                      301 The change in definition used in the compilation of the Victim Surveys does not inspire confidence. The limited extent of the sample leads to some doubt as to its validity. The proportion of unarmed robberies recorded in that survey of over 65% is not substantially different from the proportion of these recorded in the Bureau of Statistics figures for 1997 (59%), but the proportion of robberies with a firearm - less than 3% - is radically different from the Bureau of Statistics figures over many years. The Bureau information for the periods 1982/83 - 1989/90, 1991 - 1993, and 1995 - 1998 has consistently shown such offences to be a decreasing proportion of robberies as a group but the lowest proportions ever have been in 1997 (10.1 - 10.3%) and 1998 (7.7%). Given the consistency in these Bureau figures and the period involved, the proper conclusion to draw in respect of the survey figures relating to robberies with a firearm is that they are unreliable. That throws considerable doubt on the validity of the survey.
                      302 The only possible support any of the other data provide for the Survey figures is in the Criminal Court statistics which, for the period 1992 to 1995 echo to some extent the decreasing trend in the Survey figures for all robberies for the period 1990 to 1994, some time lag in the former figures being expected.
                      303 In that the Criminal Court statistics do not take into account offences not leading to a person being charged and it is not clear that they even detail the number of offences the subject of a charge, they have obvious limitations. Neither do they indicate the extent to which at the beginning and end of the periods under consideration, there may have been a backlog of cases. The Criminal Court statistics are also in absolute numbers and I confess to some surprise at the suggestion implicit in those figures that robbery has actually decreased since 1991 or 1992, despite the notorious increase in drug taking which has occurred. On the other hand, the information contained in the Report on Government Services does show decreases in some states over some periods. An example is Victoria between 1993 and 1996.
                      304 The Bureau of Crime Statistics information does have a deal of consistency about it over an extended period. In saying that I do not of course suggest that there have not been departures from trend lines and again I do not forget that direct comparison of data between periods is not possible. I have referred to the relative decrease in robberies with a firearm over three of the four periods covered by this information. There is a similar consistency, albeit in the other direction, in the data relating to robbery with a weapon other than a firearm within the periods 1982/83 - 1989/90 (except for 1982/83 itself), April 1994 - December 1995, and 1995 - 1998 and in the data relating to robbery without a weapon within the periods 1982/83 - 1989/90 (except for the first 3 of these 8 periods), April 1994 - December 1995, and 1995 - 1998. In that the information for 1991 - 1993 records that there was no statistically significant trend overall but a significant downtrend in robbery with a firearm, during that period, it indicates that there was an increase in robbery without a weapon or robbery with a weapon other than a firearm, or both, during the period. Given the length of time between 1982/83 and 1998, the exceptions I have noted do not invalidate the conclusions I have drawn as to the consistency of the information.
                      305 The difference between the information from the Bureau of Crime Statistics and that in the Report of Government Services suggests that it could not have come from the same source. There are dips in 1994 and 1996 in the armed robbery figures and these are, of course 2 out of the 5 years referred to. On the other hand the figure for 1996 is appreciably higher than for 1994 and the other three years all show an increase. The unarmed robbery figures all tend in the one direction apart from those for 1996 which are only marginally less than for 1995. Considered overall, it seems to me that the information in the Report tends in the same direction as that from the Bureau, albeit for the much shorter period of 1993 to 1997. Thus although the Report figures cast doubt on those from the Bureau, both sets of data support the conclusion that there has been a significant increase in the incidence of offending in New South Wales.
                      306 Absent other evidence, I would have been disposed to regard the Bureau of Crime Statistics as a reliable source of information in respect to statistics falling within its province such as those set out above. Having considered them, the Victim Survey data and Criminal Court statistics to which I have referred do not cause me to doubt the reliability of the Bureau statistics. If I had to choose between them, I would be disposed to regard the statistics from the Bureau as more likely to be reliable than that from the Report on Government Services but because of the limited purpose to which it is urged the statistical data should be put, I do not need to make that choice. The information from both the Bureau and the Report demonstrate that there has been a very substantial increase in the incidence of both robbery and armed robbery. The statistics from these two sources are sufficiently consistent and the periods covered by them sufficiently long that, subject to any statutory or discretionary restraints, the Court should approach the issues before it on that basis.
                      307 At the risk of repetition, the information from the Report indicates that in New South Wales between 1993 and 1997 there was an increase in armed robbery incidents per 100,000 population from 38.6 to 79.3 - 105%. If the average of 1996 and 1997 is compared with the average of 1993 and 1994, the increase is 70%. The increase in unarmed robbery and robbery overall is about 100%.
                      308 The information from the Bureau of Crime Statistics indicates-
                      Between 1983/84 and 1989/90, or between the three year period 83/84 - 85/86 on the one hand and the three year period 87/88 - 89/90 on the other, the incidence of armed robbery increased by about 13%.
                          In the period 1991 to 1993, in robbery overall there was no statistically significant trend albeit robbery with a firearm decreased by about 30.7%.
                          From the first to the last 9 months of the period April 1994 to December 1995, robbery with a weapon other than a firearm went up by 27.3%. The trend for robbery with a firearm was neither upward or downward.
                          Between 1995 and 1998 there was an increase in armed robbery incidents per 100,000 population from 35.1 to 83.6 - 138%. The percentage increase is 188% if robbery with a weapon not a firearm is the category considered.
                      309 The size of the increase between 1995 and 1998 is staggering. Given that the period covered by the figures for those years is only four years and the change is so much greater than that apparent for earlier periods, I am disinclined to rely on that data as showing reliably in quantitative terms, a long term trend. However, when regard is had to the fact that the New South Wales statistics for three of the four periods mentioned show a substantial increase in the incidence of armed robbery during those periods, and the Government Services report shows the same for the period it covers, both for New South Wales and Australia generally, I am of the view that one should accept that the offence has become far more common in this State, not merely in absolute terms but also per head of population. It must be remembered that insofar as the percentage increases to which reference has been made are accepted, they need to be compounded if one is considering the change since about 1983.
                      310 I would add that, given the increasing usage of heroin and other drugs in the community and the need of most addicts to fund their addiction by crime, the conclusion that there has been a substantial increase in the incidence of armed robbery is not surprising.
                      311 It was not suggested on behalf of the respondents to the appeals that resort to statistical material of the above kind was other than legitimate and no other inconsistent data was advanced. Certainly, the reliability of the data was criticised but I have dealt with that topic. The Court’s attention was directed to the victimisation rates for 1996 and 1997 of 0.4 and 0.3% respectively and it was submitted that these showed that armed robbery was a reasonably rare offence. However these figures indicate that the average chance of being robbed during a 50 year adult life span is between 15 and 20%. It is not obvious that that prospect should be regarded with equanimity.
                      312 The increased incidence in armed robbery, particularly where the weapon is not a firearm, justifies a review of the level of sentencing that has applied hitherto and thus a guideline judgment.
                      Evidence as to the Impact on Victims
                      313 Turning to another topic, I take the view that the evidence as to the impact of robbery on victims generally is a matter to which the Court may have regard either as a matter of judicial notice or as an “other matter” within s29(1) of the Criminal Procedure Act and a matter the Court is entitled to take into account on the question of sentence in the individual cases. It is of the same nature as the material the Court was referring to in R v Broxam (unreported, CCA, 3 April 1986):-
                      “The prevalence of the offence in respect of which the applicant was sentenced is a matter of grave social concern… It is now, well recognised that the victims of such armed robberies or attempted armed robberies may suffer devastating psychological damage consequent upon the trauma necessarily involved with the commission of such offences. In these circumstances sentencing judges would be recreant to the trust which the community places in them, if they were not to impose sentences consistent with the seriousness and prevalence of such offences.”
                      314 The passage was cited with approval in R v Bell (unreported, CCA, 11/5/93) and R v Stefanovski (unreported, CCA, 9 June 1994).
                      315 That material is relevant to sentencing standards generally and therefore to the individual cases here. Of course, any evidence as to the impact on the particular victims is also relevant whether that impact be negligible or major. But if there is no particular evidence, it does not seem to me that the Court must proceed on the basis that there has been no impact or that it is not entitled to inform itself as to the possible impact.
                      316 The full terms of Section 29A of the Criminal Procedure Act 1986 are set forth in paragraph 56 of the judgment of the Chief Justice. Sub-section (1) allows the Court to take into account “evidence or other matters”. Sub-section (2) provides that the “Court must not increase a sentence in any appeal by reason of, or in consideration of, any evidence that is used by the Court in giving a guideline judgment in the appeal and that was not given at the trial”. Normal principles of statutory construction lead to the conclusion that there is no prohibition in sub-section (2) in using “other matters” to increase a sentence. Within this expression would be the material referred to in Veen v R (No 2) (1987-88) 164 CLR 465 at 473 and that to which this Court had regard in R v Hollocoglu (1991) 29 NSWLR 67.
                      The Guideline for New South Wales
                      317 On the question of what the guideline for New South Wales should be, I agree with Spigelman CJ that the category of case to which His Honour referred is sufficiently common for the purposes of setting a guideline. However, in my view the guideline should be set at a level higher than that proposed by the Chief Justice. It should be set so as to include a total term of penal servitude of 5 years and, except in unusual circumstances, a minimum term of 3 years even when special circumstances within Section 5 of the Sentencing Act 1989 (NSW) are found. Absent special circumstances, the minimum term would be, as required by that section, not less than 3¾ years.
                      318 A minimum term should be included in the guideline because inadequacy in the length of such terms is one of the deficiencies in past sentencing practice which is apparent in the review of the Judicial Commission sentencing statistics referred to in the judgment of the Chief Justice. Given the frequency with which special circumstances have been found - and there is nothing to suggest this is likely to change - and minimum terms of less than half the full term have been imposed - matters referred to in paragraph 107 of the Chief Justice’s reasons - it is likely that, absent the inclusion of a minimum term in the guideline, there will continue to be many sentences imposed for armed robbery where the total term is 4 years and the minimum term is 2 years or less. As guidelines, I regard the former figure as too low and the latter as wholly inadequate for a term which must “remain such as appropriately reflects the criminality involved” - Morrissey (unreported, CCA, 15 July 1994, cited with approval in R v McDonald (unreported, CCA, 12 October 1998).
                      319 The statutory provisions relevant for present purposes are those contained in Sections 94 to 98 of the Crimes Act 1900 (NSW). Although actions other than robbery with or without aggravating features are encompassed within these Sections, it will be convenient in the discussion which follows to refer to the term rob and its derivatives as encompassing the all of the primary activities proscribed. The sections reveal a clear pattern of seriousness in the mind of the legislature.
                      · Simple robbery carries a maximum punishment of penal servitude for 14 years - Section 94.
                      · Robbery in circumstances of aggravation, or in company, or being armed with an offensive weapon, or accompanied by the malicious infliction of actual bodily harm carries a maximum punishment of penal servitude for 20 years - Sections 95 and 97(1).
                      · Robbery accompanied by:-
                          (a)Wounding - Section 96, 98,
                              (b) The infliction of grievous bodily harm - Section 96, 98, or
                              (c) Being armed with a dangerous weapon - Section 97(2),
                              increases the maximum penalty to one of 25 years.
                      320 (The inclusion by s4 of a dangerous weapon within the definition of “offensive “ weapon, if the definition applies throughout these provisions, does complicate the orderly pattern otherwise revealed. However, it does not affect the substance of the above picture.)
                      321 Commonly the objective circumstances which may be involved in offences against this group of sections may be considered as falling within four areas, viz. the amount of money or the value of the items involved, the premeditation and degree of planning on the part of an offender, the offender’s actions towards his victim or victims and fourthly the impact of those actions on the victim or victims. It is appropriate that, to discourage such activity, the law impose high sanctions on those who participate in criminality for high stakes. Those whose criminality extends over a greater period and displays deliberateness and planning can, all other things being equal, anticipate the duration of their criminal intent and actions to be reflected in any sentence imposed.
                      322 In considering the actions of an offender, it should be recognised that robbery is a denial of one of the elementary freedoms on which our society is based. Robbery involves the imposition of the will of the offender by force or the threat of force on the rights and ability of the victim to go about his own affairs in his own way. Some actions on the part of offenders, e.g. being in company, being armed, the malicious infliction of actual bodily harm, find express recognition in the statement of the offences. Others are also relevant. Thus conduct calculated to cause terror rather than merely the fear necessarily involved in an offence is an aggravating circumstance. Relevant also are the risks created by an offender - Readman 47 A Crim R 181 at 185, Lane (unreported, CCA, 3 November 1995). A consideration of the cases demonstrates that not all victims meekly acquiesce in a robber’s demands. Commonly a robbery occurs in circumstances charged with emotion and experience within and without the courts shows that there is but a fine line between a threat and a weapon on the one hand and injury or death on the other. Although an offender who crosses that line may lay him or herself open to further charges, it is important that the courts impose real sanctions to discourage the creation of the risk.
                      323 Some impact on the victim is specifically referred to in the statutory provisions. But it is also important to bear in mind that it is an essential element in every robbery that the victim will be put in fear of an offender and of what that offender might do. After all it is the engendering of that fear which it is the offender’s intention and by which he is able to carry out his subjection of the victim or his will.
                      324 In many cases the fear will be of imminent death with the concomitant loss of all that life holds and everything the victim holds dear. Experience of life and experience within the courts demonstrates that not all persons readily recover from such traumatic events. In other cases the fear may not be extreme and in not all cases will it continue to have an impact or significant impact on the victim after the threat causing it is removed. Not all people react to circumstances of stress in the same way. However, as I indicated above when referring to R v Broxam, R v Bell, and R v Stefanovski, the potential devastating psychological damage consequent on armed robberies is something of which the Courts have taken judicial notice over many years.
                      325 By definition, the category of offence to which the guideline in this case is directed falls towards the bottom end of the scales of amount and premeditation. That is not to suggest these aspects are of no significance. The loss of $1,000 to a small shopkeeper may be more serious to him than the loss of $1M to a bank. The premeditation will rarely be of less than some hours duration and, even if the selection of the victim has not occurred long before the commission of an offence, the intention to commit, and preparation for, the offence will often be considerably longer than this.
                      326 The category of offence does not fall so low when the other aspects to which I have referred are considered. The imposition on a shopkeeper is no different in kind from the imposition on a bank. The imposition on, fear engendered in, and the emotional or psychological reaction of the person threatened does not depend on whether they are a shopkeeper, bank teller or jewellery store owner. The risks of harm may be less with a knife than a gun, though this is not necessarily so and both can be fatal. Much will depend on often unpredictable actions of the victim and reactions of an offender who may be desperate for money and the drugs he can obtain with it. Indeed resistance by an uninsured shopkeeper faced with the loss of his own monies may be more likely than opposition from trained staff in larger organisations though one must recognise that in the latter situation there may be more persons exposed to danger.
                      327 The commencing point in any consideration of an appropriate sentence for a statutory offence is the statute itself. As was said in another context, “In determining the proper penalty … the fundamental consideration is rather the degree by which, having regard to the maximum penalties provided by the Act in question, the Respondent’s conduct would offend against the legislative objective of suppressing the illicit traffic in the prohibited drug” - R v Peel (1971) 1 NSWLR 247 at 262.
                      328 Having regard to the terms of the legislation to which I have referred and the nature of armed robbery, an offence which comes within a “worst case” category for which 20 years is the maximum penalty will commonly involve a large amount of money, a deal of premeditation, imposition on, and the risk of harm to, a number of people and serious psychological harm to at least one. It is likely it will involve “company” and/or an offensive weapon. The offender will commonly, though not necessarily, be a repeat offender. If one is considering a worst case for which 25 years is the maximum penalty, the offence is likely to involve also a dangerous weapon and the injury may be grievous bodily harm. Clearly, the category of case for which the guideline is intended falls a long way short of these sorts of offences. But so do the penalties I propose fall a long way short of 20 or 25 years. Although I do not suggest that anything shorter is insignificant, in absolute terms 5 years is but one quarter and 3 years is but 15% of the 20 year maximum. Terms of 3 and 5 years are shorter than the time spent by many people in tertiary or other training or in jobs which are of no long term significance to them. The periods are relatively short when compared with an adult life span. I said in R v Spiteri [1999] NSWCCA 3 that I regard 10 years imprisonment as more than twice as severe as a 5 years sentence The longer period is likely to have an impact on an offender’s life in terms of wife, children, job prospects and the like from which he may well never recover. Those observations apply a fortiori to a sentence of 20 or 25 years.
                      329 In my view nothing less than the terms I proposed is appropriate as a norm or guideline for a premeditated offence of the nature of that described by the Chief Justice and which can be expected to place the victim, and possibly other members of the public, both in fear and at risk of serious injury or death and which is calculated, whatever the offender may intend, to cause significant psychological or other problems in at least part of the population. However small the amount taken, however short may have been the premeditation in a particular case, and however well an individual victim may have coped with the imposition on him or her, society has a great interest in protecting its members from the incidents necessarily attendant upon any armed robbery and in reducing the risks attendant on repetition of armed robberies, even small armed robberies, in the future. The punishment imposed must be such as to amount to a substantial deterrent.
                      330 The statistics referred to when I was dealing with the topic of whether there should be a guideline judgment indicate, as I have said, that the offence of armed robbery has become far more common. The correct approach of the Court in that situation is set forth in the judgement of King CJ (with whom the other members of the court agreed) in R v Dube (1987) 46 SASR 118, quoted by the Chief Justice at paragraph 151 of his Reasons, viz. to increase the level of punishment. The increased incidence of the offence argues strongly that the guideline I propose is indeed insufficiently severe. However, as that guideline is a significant increase on the level of sentencing which has prevailed in the past, I am not persuaded that the guideline should be set at a level higher than I propose until one sees the result. After all, a significant number of offenders will receive a higher sentence than I have indicated because of the presence of aggravating circumstances such as a bad prior record and the commission of an offence while on conditional liberty.
                      Drug Addiction
                      331 I turn then to the topic of the significance of drug addiction in the determination of an appropriate penalty. On this topic I agree with the Reasons of the Chief Justice and Wood CJ at CL. In so doing, I should expressly acknowledge that I retreat from the remarks I made in R v Murray (unreported, CCA, 7 November 1995), quoted by Wood J. I am persuaded both by the reasoning of their Honours and by the authority quoted by them that, at least as a general proposition, what I said in R v Murray was wrong.

                      IN THE COURT OF
                      CRIMINAL APPEAL

                      60559/98
                      60558/98
                      60561/98
                      60746/98
                      60596/98
                      60595/98
                      60511/98

                      SPIGELMAN CJ
                      WOOD CJ at CL
                      NEWMAN J
                      HULME J
                      SIMPSON J

                                              Wednesday 12 May 1999

                      REGINA v Paul Anthony HENRY
                      REGINA v Stephen Anthony BARBER
                      REGINA v Hoai Vinh TRAN
                      REGINA v Troy David SILVER
                      REGINA v Theo TSOUKATOS
                      REGINA v Bill KYROGLOU
                      REGINA v John David JENKINS

                      GUIDELINE JUDGMENT


                      SIMPSON J :
                      332 I have read in draft the judgments of Spigelman CJ and Wood CJ at CL. I agree, for the reasons given by the Chief Justice, that it is appropriate that this court promulgate a guideline in relation to sentences imposed for offences against s 97 of the Crimes Act, and I agree with the guideline sentence proposed at para 165 of the draft judgment.
                      333 I would, however, with respect to the offence/offender profile outlined in paragraph 162 of the draft judgment, express a preference for the promulgation of a guideline that relates to sentencing after conviction following a trial. (This would, of course, require lengthier terms than those specified in the proposed range.) There are three reasons for my preference. The first is that not all pleas of guilty carry the same weight, depending as they do on the strength of the Crown case, and the time at which they are offered. If the starting point assumed a plea of not guilty, then the appropriate discount to reflect the true worth of the plea could be given. The second reason is that the severity of the sentence to be imposed in relation to any offence might be mitigated by numerous facts or features, of which a plea of guilty is only one. I see no reason to single out the plea of guilty from other mitigating features in the definition of the kind of offence to which the guideline applies. The third and most substantial reason concerns appearances. Although the truth is that the guideline sentence has built into it a discount representing the benefit to which an offender is entitled for pleading guilty, the appearance, when an offender is sentenced following a trial, will be of a penalty that is attributed to the exercise of the right to the put the Crown to proof. The appearance would, as I have said, be false, but would exist nonetheless.
                      334 However, as I understand that other members of the Bench do not share these concerns, I am content to concur in the guideline as proposed.
                      335 I wish to say something about the submission advanced on behalf of certain of the respondents to the effect that drug addiction when a causal factor in the commission of offences should be recognised as a significant mitigating feature.
                      336 It is a mistake, in my opinion, to regard drug addiction as a starting point. It is an end point, or a point on the way to the end, of a process. Where the process begins may vary in individual cases but it does not necessarily begin when a person decides to use a prohibited substance. It probably has its origins well before the date of the first use of drugs. In the worst, or least forgivable, cases it may have its origins in arrogance, in an antipathetical attitude to the laws of society, or in weakness of character. In other cases, I have no doubt, it has its origins in social disadvantage, poverty, emotional, financial, or social deprivation, poor educational achievement, unemployment, and the despair and loss of self-worth that can result from these circumstances or any combination of them. In this court one sometimes sees cases in which drug taking stems from sexual assault or exploitation, sometimes committed when the person who turns to drugs, and who comes before the court, is very young, and sometimes the precipitating events have occurred many years before. Drug addiction is not always the disease; it is, as often as not, a symptom of social disease.
                      337 Drug addicts do not come to their addiction from a social or environmental vacuum. This court should not close its eyes to the multifarious circumstances of disadvantage and deprivation that frequently precede and precipitate a descent into illegal drug use. I do not suggest for a moment that all drug users fall into this category. It is because some do and some do not that I believe rigid rules about the impact on sentencing of drug dependency on sentencing cannot be laid down.
                      338 I cannot accept that the blameworthiness of one drug taker is (even excepting that small number of individuals who begin drug taking with medically prescribed drugs) always to be treated as being at the same level as the blameworthiness of the next. Nor can I accept that the exercise of free choice in the use of drugs is always of equal dimensions. It is not every decision to use drugs that can properly or fairly be characterised as a decision made in the exercise of free choice. The will of an individual can be overborne or undermined, not only by acts of another person, but also by the pressure of circumstances. I do not accept that most drug offenders are truly exercising free will when they choose the degradation, despair, criminality and cycle of imprisonment that can follow the initial use of illegal drugs. The circumstances that propel the offender to the use of drugs are often, if not usually, beyond his or her control. They may or may not be combined with a vulnerable personality or even a weakness of character. Many drug offenders have not had the life experiences or the normal developmental path that permit a conclusion that the decision to take drugs was a decision made in the exercise of a free choice in the sense in which that phrase is ordinarily understood.
                      339 An analogy can legitimately and properly be drawn with the principles relating to the sentencing of aboriginal offenders, collected and stated by Wood J (as he then was) in R v Fernando (1992) 76 A Crim R 58 at 62 - 63. As I understand the remarks on sentence in that case, his Honour observed that aboriginality alone does not excuse serious crime nor operate as a mitigating factor. However, since it is often associated with other circumstances of disadvantage or deprivation, aboriginality may explain or throw light on the particular offence or the circumstances of the offender. It is, in truth, not aboriginality, but the life experiences too commonly associated with aboriginality that are relevant to the sentencing decision. Just as it is properly said that, while drunkenness is not normally an excuse or mitigating factor, where alcohol abuse reflects the socioeconomic circumstances and the environment in which an offender has grown up, that fact can and should be taken into account as a mitigating factor (Fernando, principle E, p 62), so also can it properly be said that drug abuse may reflect the socioeconomic circumstances and the environment in which another offender has grown up, and that may equally then be taken into account as a mitigating factor.
                      340 Wood J went on to recognise the great social difficulties sometimes faced by aboriginal communities where poor self image, absence of education and work opportunity and other demoralising factors have placed heavy stresses on them, reinforcing their resort to alcohol and compounding its worst effects. So, too, it is with some drug offenders. The passage from which the above is drawn can readily be adapted to the situation of some drug offenders. Where great social difficulties, poor self image, absence of education and work opportunity and other demoralising factors have placed heavy stresses upon drug offenders, causing or reinforcing their resort to drugs, then it is appropriate for this court to give the same recognition to those antecedent circumstances.
                      341 A sentencing judge always has a delicate balancing task to perform, taking into account the demands of the community for retribution, deterrence both general and specific, and the interests of the same community in the rehabilitation of an offender. The exercise may call for an examination of the circumstances that led the offender to drug use, addiction and crime. All the circumstances that precipitate the use of drugs are relevant to the evaluation of moral culpability that is essential to the sentencing process.
                      342 It would, therefore, be too simplistic to lay down a principle that addiction either is, or is not, a mitigating circumstance in the sentencing of offenders convicted of drug related crime. I much prefer the approach encapsulated in paragraphs 270 - 272 inclusive of the draft judgment of Wood CJ at CL, favouring the “individualised justice” to which Mahoney ACJ referred in R v Lattouf (unreported, NSWCCA, 12 December 1996).
                      343 While, on this approach, it might be possible to argue that moral culpability for the offence is diminished, suggesting a more lenient rather than a more severe penalty, the counter balance is that, absent evidence of positive prospects of a cure of the drug addiction, a finding that the offence is unlikely to be repeated (itself ordinarily seen as a factor in favour of a lesser sentence) would be unavailable. Absent positive evidence of real rehabilitation, the diminution in moral culpability which results from drug addiction originating in the kind of social or familial deprivation to which I have referred cannot result in a corresponding diminution in sentence. Evidence of rehabilitation is the key to the impact evidence of drug addiction might have in the sentencing process. This view I perceive to be largely in accord with the direction taken by the legislature in the establishment of a Drug Court, specifically created to divert drug offenders with realistic prospects of rehabilitation (other than those involved in violence) from the criminal justice system: Drug Court Act 1998 (NSW) S 5(2)(b); para 4 of the draft judgment of Wood CJ at CL.
                      344 Where circumstances such as those I have mentioned (or others equally deserving of compassion) have been the foundation for the drug addiction, and part of the causal chain leading to the commission of crime, then it would be appropriate, in my view, for the rehabilitative aspects of sentencing to assume a more significant role than might otherwise be the case. In an appropriate case, rehabilitation might outweigh other sentencing factors. In order for those circumstances to provide a reason for reduction of sentence, however, there would need to be strong evidence of real progress towards actual rehabilitation. I would not wish to be understood to be saying other than that leniency of the kind to which I refer depends heavily upon demonstrated (as distinct from theoretical) rehabilitative prospects.
                      345 The authorities cited by Spigelman CJ and Wood CJ at CL are almost unanimous in holding that drug addiction of itself does not operate as a mitigating circumstance. When the words “of itself” are emphasised, I respectfully agree with the proposition repeatedly stated; but, in any event, the line of authority is so compelling and so sustained that, even sitting as one of a bench of five, I would be reluctant to participate in a decision departing from it. It is the role of the legislature (or the High Court) to alter a such a long standing principle. What I have said above is consistent with that line of authority . Nowhere is it held that it is inappropriate, in sentencing drug offenders, to take into account either the circumstances that gave rise to the drug addiction or demonstrated rehabilitation.
                      346 In all the mass of material put before this court to support the argument that drug addiction should be treated as a mitigating factor, there was nothing to inform the court, statistically or otherwise, of what causes drug addiction, or its socioeconomic background. What I have said, therefore, is somewhat impressionistic, much of it drawn from experience on this court, and some of it assumptions based upon more general reading. It is, I believe, largely uncontroversial. It is not intended as a recitation of the most universal or common circumstances pre-dating or causing drug addiction, but of some familiar patterns.
                      347 I have not lost sight of the material to which the court was taken concerning the effect of armed robberies on their victims. Obviously, the offence is serious, commonly has drastic consequences for victims, and must be punished accordingly.
                      348 I agree that the principles stated by Wood CJ at CL in paragraph 273 of his draft judgment are the principles that emerge from the authorities. To them, I would add that not all drug addicts are necessarily to be taken to have brought themselves to addiction and criminality with the same degree of foresight, determination, informed decision making, and consequent moral culpability. To impute those qualities to every decision to take drugs is to ignore the important starting point, the reasons underlying the initial drug use.
                      349 I entirely agree with the proposition that the bare fact that an offence is motivated by a need for money to support a drug habit does not, alone, mitigate the offence or operate to reduce the sentence to be imposed.
                      350 Counsel who appeared for some respondents also argued, by analogy with the sentencing of mentally abnormal offenders, that, in sentencing drug addicted offenders, it may be appropriate to afford less weight to the principle of general deterrence. This argument cannot be accepted. In considering general deterrence it is necessary to consider the class or pool of individuals at whom deterrence is directed. In s 97 offences, that class or pool is the class of people who commit, or are tempted to commit, the offence of armed robbery - many of whom, as the materials show, are also motivated by the need for money to feed drug addiction. These are people whose will not to commit crimes is already weakened by the drug dependence. That means an even stronger message must be conveyed. Far from diminishing the importance of general deterrence, the fact that the population in which deterrence is necessary is unlikely to be deterred other than by firm action on the part of the courts speaks for greater rather than lesser emphasis on that principle.
                      351 Usually, and certainly in relation to armed robbery, general deterrence is a very significant factor in the sentencing process. So is specific deterrence and so is punishment. The point I wish to make is that where a combination of two circumstances exists, then general deterrence, and other sentencing objectives such as retribution, may yield to rehabilitation. The two circumstances are:
                      352 (i) a background to the addiction that may explain the offender’s decision to use drugs, and diminish his or her moral culpability for that decision;
                      353 (ii) demonstrated progress towards rehabilitation.

                      354 The second is, to my mind, essential before general deterrence can be seen to give way to rehabilitation.
                      355 Where those two circumstances coincide, then the interests of the community may well be better served by the imposition of a penalty that leans towards furthering the rehabilitative process at the expense of the punitive and even the deterrent objectives of sentencing.
                      356 I do not understand this approach to be significantly different from the recognition given by Wood CJ at CL to the need for sentencing to be sufficiently flexible to take account of all relevant circumstances (paragraphs 270 - 2719). It is consistent with many authorities in this court and other jurisdictions: see, eg R v Lattouf (unreported, NSW CCA, 12 December 1996); R v Hayes (1987) 29 A Crim R 452 at 457, 472; R v Eager (unreported, NSW CCA 1 November 1995); R v Crotty (unreported, NSW CCA 28 February 1994); R v Fabian (1992) 64 A Crim R 365 at 372, 378, 380; R v Molina (1984) 13 A Crim R 76 at 77 (Federal Court of Australia) (quoted with approval in R v Ellis (1993) 68 A Crim R 449 by Kirby P, who, however, dissented in the result in that case); see also the remarks in Molina in relation to the offender there before the court at p 79.2; R v Dowie (1989) 42 A Crim R 234 at 247 (Court of Criminal Appeal, Tasmania); R v Osenkowski (1982) 30 SASR 212 at 212-3 (Supreme Court of South Australia in Banco); R v Halewyn (1984) 12 A Crim R 202 at 205-6 (Court of Criminal Appeal, Victoria).
                      **********


                    | Previous Page | Back to CASELAW NSW | Top of Page |

                    Disclaimer | Webmaster | Feedback
                    spacer image
                    © Copyright 1999 Crown Copyright
                    All material is reproduced by permission of the Crown but does not purport to be the official or authorised version. Downloading, copying or printing of materials in this database for personal use, or on behalf of another person, is permitted. Downloading, copying or printing of material from this database for the purpose of reproduction or publication (in whole or in part) for a fee is not permitted without express authorisation.