Tillman v Attorney General for the State of New South Wales
[2007] NSWCA 327
•26 November 2007
NEW SOUTH WALES COURT OF APPEAL
CITATION: Tillman v Attorney General for the State of New South Wales [2007] NSWCA 327
FILE NUMBER(S):
40537/07
HEARING DATE(S): 27/09/07
JUDGMENT DATE: 26 November 2007
PARTIES:
Kenneth Davidson Tillman (Claimant)
Attorney General for the State of New South Wales (Opponent)
JUDGMENT OF: Mason P Giles JA Ipp JA
LOWER COURT JURISDICTION: Supreme Court - Common Law Division
LOWER COURT FILE NUMBER(S): SC 11830/07
LOWER COURT JUDICIAL OFFICER: Bell J
LOWER COURT DATE OF DECISION: 18/06/07
LOWER COURT MEDIUM NEUTRAL CITATION:
[2007] NSWSC 605
COUNSEL:
A Haesler SC/S Beckett (Claimant)
L Babb/A M Mitchelmore/J Caldwell (Opponent)
SOLICITORS:
Legal Aid Commission of NSW (Claimant)
I V Knight, Crown Solicitor (Opponent)
CATCHWORDS:
STATUTORY INTERPRETATION – Crimes (Serious Sex Offenders) Act 2006 (NSW) – continuing detention orders under s 17(3) – construction of the term “likely” – whether, in the light of the standard of proof in s 17(3) and the common law presumption against the infringement of personal liberty, the term “likely” is used in s 17(3) as meaning “more probable than not”
COURTS AND JUDICIAL SYSTEM – principle of comity – whether an intermediate appellate court of one Australian jurisdiction should, on grounds of comity, follow a decision of an intermediate appellate court of another Australian jurisdiction where the issue under consideration involves the interpretation of legislative provisions that are identical or substantially similar. D
LEGISLATION CITED:
Civil Liability Act 2002 (NSW)
Criminal Code Act 1924 (Tas), s 157(1)
Serious Sex Offenders Monitoring Act 2005 (Vic), s 11(1)
Charter of Human Rights and Responsibilities Act 2006 (Vic)
National Security (Landlord and Tenant) Regulations 1941 (Cth)
Crimes (Serious Sex Offenders) Act 2006 (NSW), ss 3, 11, 16, 17(1), (2), (3), (4)(c), (d), (i), 18(1)(b), 19
CASES CITED:
Attorney General for the State of New South Wales v Gallagher [2006] NSWSC 340
Attorney General for the State of New South Wales v Tillman [2007] NSWSC 605
Attorney General for the State of New South Wales v Tillman [2007] NSWCA 119
Attorney General for the State of New South Wales v Winters [2007] NSWSC 1071
Australian Consolidated Press Ltd v Uren [1969] 1 AC 590
Australian Securities Commission v Marlborough Gold Mines Ltd (1993) 177 CLR 485
Beck v The Queen [1984] WAR 127
Boughey v The Queen (1986) 161 CLR 10
Brennan v Comcare (1994) 50 FCR 555
Camden Park Estate Pty Ltd v O’Toole (1969) 72 SR (NSW) 188
Cook v Cook (1986) 162 CLR 376
Darkan v The Queen (2006) 227 CLR 373
Esso Australia Resources Ltd v Commissioner of Taxation (1999) 201 CLR 49
Farah Constructions Pty Ltd v Say-Dee Pty Ltd (2007) 81 ALJR 1107
Fernando v Commissioner of Police (1995) 36 NSWLR 567
Lange v Australian Broadcasting Corporation (1997) 189 CLR 520
Macquarie Bank Limited v Fociri Pty Ltd (1992) 27 NSWLR 203
Ogden Industries Pty Ltd v Lucas [1970] AC 113
Parker v The Queen (1963) 111 CLR 610
Plaintiff S157/2002 v The Commonwealth of Australia (2003) 211 CLR 476
Public Transport Commission of New South Wales v J Murray-More (NSW) Pty Ltd (1975) 132 CLR 336
R v Abbrederis [1981] 1 NSWLR 530
R v Gardiner [1981] Qd R 394
R v Lorkin (1995) 15 WAR 499
R v NZ (2005) 63 NSWLR 628
R v Parsons [1983] 2 VR 499
R v Yates (1991) 102 ALR 673
Re Bolton; Ex parte Beane (1987) 162 CLR 514
Re J & E Holdings Pty Ltd and the Corporations Law (1995) 36 NSWLR 541
Rejfek v McElroy (1965) 112 CLR 517
Sullivan and Rigby v The Queen (1987) 47 NTR 31
TSL v Secretary to the Department of Justice (2006) 14 VR 109
Viro v The Queen (1978) 141 CLR 88
Zibillari v The Queen [1981] WAR 40
DECISION:
(1) Leave to appeal granted. (2) Appeal dismissed.
JUDGMENT:
IN THE SUPREME COURT
OF NEW SOUTH WALES
COURT OF APPEAL
CA 40537/07
SC 11830/07MASON P
GILES JA
IPP JAMonday 26 November 2007
KENNETH DAVIDSON TILLMAN v ATTORNEY GENERAL FOR THE STATE OF NEW SOUTH WALES
The claimant/appellant, Kenneth Tillman, was sentenced to ten years’ imprisonment following his conviction for a serious sex offence. He completed his sentence on 18 April 2007 and was released from prison subject to an interim supervision order.
On 3 May 2007, the New South Wales Court of Appeal ordered that Mr Tillman be detained in a correctional centre for 28 days. This interim order was renewed on 29 May 2007 for a further 28 days as from 30 May 2007.
By amended summons dated 25 May 2007, the opponent/respondent, the Attorney-General for the State of New South Wales, applied for an order pursuant to s 17(1)(b) of the Crimes (Serious Sex Offenders) Act 2006 (NSW) (“the Act”) that Mr Tillman be detained in a correctional facility for five years from the date of the order. Such continuing detention orders “may be made if and only if the Supreme Court is satisfied to a high degree of probability that the offender is likely to commit a further serious sex offence if he or she is not kept under supervision and that adequate supervision will not be provided by an extended supervision order” (s 17(3)).
On 18 June 2007, Bell J ordered that Mr Tillman be detained in a correctional facility for one year.
Mr Tillman appealed from her Honour’s decision.
The issues to be determined by the Court of Appeal were as follows:
(i)whether, as submitted by the claimant / appellant, the trial judge erred by failing to find that the term “likely” in s 17(3) meant “a high degree of probability” such that the likelihood be more probable than not; and
(ii)whether the principle of comity calls for the adoption by an intermediate court of appeal of one Australian jurisdiction of a decision by an intermediate court of appeal where both courts have considered legislation in similar or identical terms.
(iii)whether the trial judge should have found, on the facts, that the degree of satisfaction required by s 17(3) was satisfied.
HELD per GILES and IPP JJA, dismissing the appeal:
In relation to (i):
Though there is much to be said for the notion that “likely” is used in s 17(3) as meaning “more probable than not” (see, for example, Attorney General for the State of New South Wales v Winters [2007] NSWSC 1071 per McClellan CJ at CL), so too the view that, by requiring satisfaction to a high degree of probability as the standard of proof, the legislature recognised the common law presumption against the infringement of personal liberty. Consequently, it would be incorrect to impose a further strict requirement of likelihood in the sense of more probable than not. Thus, the term “likely” denotes a high degree of probability, but not necessarily involving a degree of probability in excess of 50 per cent; that is, such likelihood need not be more probable than not: [88] to [90], [92].
TSL v Secretary to the Department of Justice (2006) 14 VR 109, applied.
Attorney General for the State of New South Wales v Winters [2007] NSWSC 1071, considered.
Per MASON P, also dismissing the appeal:
The term “likely” should be given a meaning of “more probable than not”: [8] to [13].
As to the standard of proof in s 17(3), the Act’s requirement of judicial satisfaction to a high degree of probability should be read in the light of the principle that the degree of satisfaction for which the civil standard of proof calls may vary according to the gravity of the fact to be proved. Interpreting the standard in this manner circumvents the semantic dispute that “probability” and “likelihood” can never be equated, and meets the suggestion that the reference in s 17 to probability to a high degree compels one to interpret “likely” as meaning something less than 50 per cent so as to ensure the Court does not have to apply conflicting degrees of probability: [14] to [17].
Rejfek v McElroy (1965) 112 CLR 517, referred to.
Per GILES and IPP JJA:
In relation to (ii):
The approach taken by Callaway AP in TSL v Secretary to the Department of Justice (2006) 14 VR 109 should be adopted in the present case for the reason that an intermediate appellate court of one Australian jurisdiction should, on the ground of comity, follow the decision of an intermediate appellate court of another Australian jurisdiction where the relevant issue concerns the interpretation of statutory provisions that are identical or substantially similar (unless the Court is convinced that the earlier decision was clearly wrong or if considerations of justice require the Court not to apply it): [106] to [110].
Fernando v Commissioner of Police (1995) 36 NSWLR 567, applied.
R v NZ (2005) 63 NSWLR 628; Sullivan v The Queen (1987) 47 NTR 31; R v Lorkin (1995) 15 WAR 499, referred to.
In the present case, s 17(3) of the Crimes (Serious Sex Offenders) Act is sufficiently close to the terms of s 11(1) of the relevant legislation in TSL, the Serious Sex Offenders Monitoring Act 2005 (Vic), as to require the Court to follow the Victorian Court of Appeal decision in TSL: [73].
Per MASON P:
Though there does exist a rule requiring Australian intermediate appellate courts not to depart from decisions of intermediate appellate courts of other Australian jurisdictions on the interpretation of Commonwealth legislation, uniform national legislation or the common law of Australia unless convinced of plain error, it has not been expressed by the High Court of Australia in such broad terms as to encompass deference to decisions of Australian intermediate appellate courts on the interpretation of identical or substantially similar statutory provisions: [25] to [26].
Australian Securities Commission v Marlborough Gold Mines Ltd (1993) 177 CLR 485; Farah Constructions Pty Ltd v Say-Dee Pty Ltd (2007) 81 ALJR 1107, referred to.
Though it is highly desirable that there be consistency between the decisions of Australian intermediate courts of appeal, as judges have repeatedly held, this is not a legal test for determining whether and in what circumstances a court may depart from uniformity. In the present case no categorical rule of precedent (similar to those which were expounded in Marlborough and Farah) obliging an intermediate appellate court not to depart from decisions of other intermediate appellate courts when with respect to identical or substantially similar state or territory legislation applies: [46].
Per MASON P, GILES and IPP JJA
In relation to (iii):
On either construction of s 17(3), Bell J’s examination of the facts of the case shows that the degree of satisfaction required by s 17(3) was established: [52], [94].
Orders
(a) Leave to appeal granted.
(b) Appeal dismissed.
**********
IN THE SUPREME COURT
OF NEW SOUTH WALES
COURT OF APPEAL
CA 40537/07
SC 11830/07MASON P
GILES JA
IPP JAMonday 26 November 2007
KENNETH DAVIDSON TILLMAN v ATTORNEY GENERAL FOR THE STATE OF NEW SOUTH WALES
Judgment
MASON P: Where there is an application for a continuing detention order under the Crimes (Serious Sex Offenders) Act 2006 the Court is presented with three choices under s17(1), an extended supervision order, a continuing detention order, or dismissal of the application.
The minimum criteria for an extended supervision order are set out in s17(2) of the Act. Those for a continuing detention order are set out in s17(3). It is convenient to set out subsections (2) and (3):
(2)An extended supervision order may be made if and only if the Supreme Court is satisfied to a high degree of probability that the offender is likely to commit a further serious sex offence if he or she is not kept under supervision.
(3)A continuing detention order may be made if and only if the Supreme Court is satisfied to a high degree of probability that the offender is likely to commit a further serious sex offence if he or she is not kept under supervision and that adequate supervision will not be provided by an extended supervision order.
In each instance, the Supreme Court’s power is enlivened only if the Court is satisfied to a high degree of probability that certain matters apply. The words following “probability” in each subsection define what the Court has to be satisfied about.
I shall use the expression “highly satisfied” as shorthand for the phrase “satisfied to a high degree of probability”. In doing so, I do not overlook the need to revisit the question as to what is conveyed by the language of probability in this context.
The Court is, in each instance, directed to be highly satisfied about the future likelihood of relevant reoffending in the sense of commission of a further serious sex offence. The predictive inquiry is specific to the particular offender. Furthermore, in each instance the Court must consider the likelihood of relevant reoffending in the context of the statutory power to order that the offender be kept under supervision.
If the Court perceives itself able to devise a regime of extended supervision that will make it unlikely that the particular offender will relevantly reoffend, then the combined effect of subsections (2) and (3) is that the lesser control is to be chosen. But if reoffending remains likely despite an extended supervision order, then the power to impose the more stringent control of a continuing detention order is engaged.
The universal requirement to construe words in context becomes even more pressing for chameleon-like words such as “likely”. Dictionaries and decisions of the High Court inform us that this word may sometimes mean “probable” in the sense of more probable than not (see the review by McClellan CJ at CL in Attorney General for the State of New South Wales v Winters [2007] NSWSC 1071 at [33]-[47]).
Here one is dealing with a predictive inquiry referable to the particular offender that implicitly addresses the time frame within which the Court’s order can operate. Is reoffending within this time frame likely unless detention or at least supervision is ordered? The likelihood of a single reoffending of the requisite nature will suffice to engage the various powers.
This understanding of the individualised nature of the inquiry indicates to me that the Court is required to address a cascading set of predictions. At each step the ultimate question is whether or not the Court is highly satisfied that a single event of reoffending is likely.
To my mind this suggests a preferred meaning for “likely” of more probable that not.
Imagine a competition where a team has to win a game in order to progress to the finals. The question whether or not it is likely that it would win involves a “yes:no” prediction that concerns itself with a single issue (winning or not winning). There is simply no room or reason for predicting that the team might get close to winning. Unless it wins, it does not get to the finals.
The subject matter of the inquiry in the present matter is similar in nature. Unless the Court is satisfied that reoffending is likely then it cannot make an order. The predictive inquiry is referable to a single future event.
The common law’s presumption in favour of the liberty of the subject underpins the predictive inquiry to be undertaken and further explains why this opaque legislation should be interpreted strictly in the sense that the available, tighter meaning of “more probable than not” should be chosen for “likely”. This is the matter to be addressed by the Court.
As to the standard of proof, the statute gives clear direction that ought not to be glossed unless it is necessary and helpful to do so. That standard is satisfaction to a high degree of probability. This obviously falls somewhere between a slight preponderance and satisfaction beyond reasonable doubt.
The common law knows only two standards of proof, the civil and the criminal, yet it recognises that “the degree of satisfaction for which the civil standard of proof calls may vary according to the gravity of the fact to be proved” (Rejfek v McElroy (1965) 112 CLR 517 at 521 per Barwick CJ, Kitto J, Taylor J, Menzies J and Windeyer J). The Act’s requirement of judicial satisfaction to a high degree of probability should be read in the light of this principle, in my view.
Viewed thus, one avoids the tension posed by the semantic argument that probability and likelihood can never be equated. This approach also meets the suggestion that s17’s reference to “probability” (to a high degree) forces one to interpret “likely” as meaning something less than 50 per cent in order to avoid the Court being faced with the conundrum of applying conflicting degrees of probability.
Section 17 does not invite the Court to pursue the will-of-the-wisp of some unidentified degree of mathematical probability. The judicial task is serious and difficult enough without this diversion. Juries are frequently instructed not to parse the expression “beyond reasonable doubt”. Judges should instruct themselves in similar terms as regards the expression “satisfied to a high degree of probability”.
This conclusion is at variance with that reached by the Victorian Court of Appeal in TSL v Secretary to the Department of Justice (2006) 14 VR 109 (TSL).
Giles JA and Ipp JA have concluded that this Court is required to follow TSL unless satisfied that it was clearly wrong. Like my colleagues, I am not convinced that the Victorian Court of Appeal’s interpretation is plainly wrong. But I do firmly prefer a different interpretation. I recognise that the Victorian legislation is relevantly indistinguishable.
Am I permitted to depart from or bound to defer to the Victorian Court’s interpretation in this matter?
The present case does not involve departure from any decision or dicta of the High Court with reference to the interpretation of this particular legislation. Nor is there any prior decision of this Court. Nor is it a situation where maintaining uniformity would deter the problems of forum shopping (cf Macquarie Bank Limited v Fociri Pty Ltd (1992) 27 NSWLR 203 at 217 per Kirby P). Nor is the Court dealing with a field of law, such as property or commercial law, in which people may have charted a course in the expectation that an earlier precedent of another jurisdiction represents the law.
The issue is not whether this Court is bound by the decision of the Victorian Court of Appeal. Clearly it is not. Nor is it whether this Court finds the Victorian precedent persuasive. The question is the more nuanced one as to the standard applicable for determining whether, having considered it, we should depart from it.
I would start from the general proposition that a judge has a duty to exercise his or her independent judgment in the interpretation of a statute (Ogden Industries Pty Ltd v Lucas [1970] AC 113 at 127 (PC). See also Brennan v Comcare (1994) 50 FCR 555 at 572 (Gummow J)).
Nevertheless, there are well established categories where a court that is not bound by an earlier decision (of itself or another court) will defer to and follow that decision, contrary to its own views, unless clearly satisfied that it is plainly wrong. This particular principle of deference reflects a proper sense of humility as well as the importance of stability and precedent in a common law system.
I acknowledge that there is a rule of precedent obliging intermediate appellate courts not to depart from decisions in intermediate appellate courts in another jurisdiction on the interpretation of (a) Commonwealth legislation or (b) uniform national legislation or (c) the common law of Australia unless convinced of plain error (Australian Securities Commission v Marlborough Gold Mines Ltd (1993) 177 CLR 485 at 492 (Marlborough); Farah Constructions Pty Ltd v Say-Dee Pty Ltd [2007] HCA 22; 236 CLR 209 at [135] (Farah)). To my understanding, the High Court has not expressed the rule in broader terms. Neither has it expounded the reasons that lie behind the existing rule, leaving a margin of inference on that topic.
The legislation under consideration in this appeal is not federal or part of a uniform national scheme. It does not involve matters of property, commercial law or even tort law (cf the Civil Liability Act 2002) where citizens of Australia may have had transactions or dealings in more than one State on similar terms, or made insurance or travel arrangements on assumptions of commonality throughout Australia.
Sex offenders in New South Wales are unlikely to have chartered their course in or out of prison on the basis of the Victorian Court of Appeal’s interpretation in TSL.
The Act calling to be construed is a New South Wales enactment addressing the consequences of past and likely future criminal activity. It involves the liberty of the subject. There is no evidence suggesting that it was enacted as part of a national scheme.
Respect for the decision of a court of co-ordinate authority in another law jurisdiction stems from more than a proper sense of humility. If an appellate court in one jurisdiction declines to follow a decision of a co-ordinate appellate court in another jurisdiction trial courts in the first jurisdiction will be forced to change their tune by ceasing to follow the other jurisdiction’s decision (if they have commenced to do so, as well they might have). This however is something that may happen even if there is no competing external precedent and it stems from (a) the duty to follow appellate decisions within a single jurisdiction and (b) the capacity of the law to refine, evolve and even change. Adherence to what is perceived to be an erroneous non-binding precedent is not a self-evident virtue to be pursued blindly in every legal situation.
There is no pattern of New South Wales decisions following TSL. This case is in fact the first that was determined in New South Wales on a final basis.
There will undoubtedly be circumstances in which it might be prudent for an intermediate appellate court to hold back from taking a step that could be left up to the High Court of Australia to take, if so minded, in resolving disagreements between intermediate appellate courts and/or the principled development of the law. In matters involving the grant of leave to appeal such considerations may well be germane to a proper exercise of a judicial discretion.
But the judge of an intermediate appellate court determining an appeal remains bound by the judicial oath to do justice according to law. In a proper case, that duty will require departure from non-binding precedents if the judge discerns that legal principles of higher authority require such departure. Each individual judge is obliged to pursue justice, not prudence, even though it may turn out that his or her views do not obtain the assent of appellate colleagues or are overturned upon appeal to the High Court.
I am aware of some decisions suggesting that what I have described as the principle of deference extends to decisions of intermediate appellate courts referable to the interpretation of substantially similar statutory provisions.
In Fernando v Commissioner of Police (1995) 36 NSWLR 567 (Fernando) this Court was asked to follow a decision of the Full Court of the Supreme Court of South Australia concerning State legislation dealing with the taking of blood samples.
Clarke JA said (at 587) that there was an abundance of authority in support of the proposition that considerations similar to those discussed in Marlborough applied to the interpretation of substantially similar statutory provisions in different States. At 589 he said that the relevant principle should be taken as settled in the light of those decisions. With respect to his Honour, the authorities cited, while expounding the obvious virtues of consistency, on my reading all involved Commonwealth legislation with the exception of Camden Park Estate Pty Ltd v O’Toole (1969) 72 SR(NSW) 188 (Camden Park Estate) (discussed below). His Honour nevertheless stated his support for what he described as “a most important principle … [of] control … necessary to prevent the unbridled development of divergent principles”. That principle is stated in categorical terms in the passage (at 589-90) quoted at length by Giles JA and Ipp JA at [107].
In Fernando, Priestley JA described the “rule of comity” as being a rule “pursuant to which [the Court] usually follows the decisions of courts of like position in other Australian jurisdictions on similar points” (emphasis added). Priestley JA considered the rule of comity to be “a sound one for most purposes” (at 584), but his Honour did not refer to the authorities or discuss the matter at any length. Nor did he explain why the usual rule was not sound for all purposes.
Powell JA (at 593) cited Camden Park Estate as authority for the proposition that it was “highly desirable” that there be conformity of decisions throughout Australia even with respect to legislation other than a Commonwealth statute or uniform national legislation. His Honour also pointed out specifically that the legislation involved in Camden Park Estate had stemmed from the Commonwealth National Security (Landlord and Tenant) Regulations.
While acknowledging that others might disagree with my reading of Priestley JA and Powell JA, I do not understand their Honours’ reasons as endorsing the categorical rule of precedent stated by Clarke JA that is now adopted by Giles JA and Ipp JA as binding in the present case.
The topic is addressed in the recent decision of the New South Wales Court of Criminal Appeal in R v NZ (2005) 63 NSWLR 628, [2005] NSWCCA 278 (NZ). Howie J and Johnson J, with whom Wood CJ at CL and Hunt AJA agreed, refer at 665[158] to Camden Park Estate and Fernando before stating (at 666[160]) that:
The rule of comity referred to in Fernandoappears to have application where the court is considering a decision of another Australian intermediate appellate court that has interpreted an identical or substantially similar provision in a State statute. (emphasis in original except for portion underlined)
Their Honours construed the various judgments in Fernando as speaking with one voice because they effectively define the “rule of comity” globally by reference to the key passages in the respective judgments of Priestley JA, Clarke JA and Powell JA (see at 665[159]-[160]). This is a conclusion that, with respect, I do not share for reasons already stated. I would nevertheless pause long and hard before departing from a binding rule emerging from a recent decision of the Court of Criminal Appeal. I need to go no further than stating that the reasoning in NZ that discerned a single “comity principle” to have emerged from the three judgments in Fernando leaves such a leeway of choice that the outcome in the present case is not controlled by Clarke JA’s particular reasoning in Fernando.
In NZ Spigelman CJ dissented on the relevant point, but did not discuss the comity principle.
In Sullivan and Rigby v R (1987) 47 NTR 31, the Supreme Court of the Northern Territory - Court of Criminal Appeal followed Beck v R [1984] WAR 127, a decision of the Court of Criminal Appeal in Western Australia, stating (at 36):
In the interests of comity and conformity of decision and for the reasons stated by the Court of Appeal in Camden Park Estate Pty Ltd v O’Toole (1969) 72 SR(NSW) 188 at 190, we adopt the approach in Beck v R, as there are no relevant statutory differences between the Western Australian and Territory provisions.
In Camden Park Estate Pty Ltd this Court said (at 190):
It is highly desirable that there be conformity of decision between States where legislative provisions are identical. Such conformity can be achieved by High Court decisions. It can also be greatly assisted by dicta of the High Court. A combination of High Court dictum and Victorian decision is in our view sufficient to warrant this Court in declining to follow the earlier New South Wales decision, whatever may be its view upon the correctness of the various conflicting decisions.
It can be seen that the principle stated turned, in part, upon the existence of High Court dicta pointing the way forward (see also Farah at [135]). Camden Park Estate also involved the phenomenon of this Court being asked to depart from its own earlier authority in such a context.
I do not think that Camden Park Estate is authority prescribing a categorical deference rule in the present context, where all that is involved is a disagreement between courts of coordinate jurisdiction with respect to identical legislation.
I agree that it is “highly desirable” that there be conformity of decision as between different States, but this is not a legal test for determining whether and in what circumstances a considered departure from uniformity may occur. While accepting the possibility that I may have misread the remarks of Priestley JA and Powell JA in Fernando, I remain of the view that no categorical rule of precedent similar to that expounded in Marlborough and Farah applies in the present context.
The decision of Giles JA and Ipp JA in the present matter will, however, establish such a rule. I respectfully dissent.
My colleagues hypothesise that the public would be entitled to ask why the same words of a statute have different meanings in different States or Territories. Rhetorical arguments are seldom compelling, so I shall content myself with the riposte that the public within one State might reconsider their instinctive answer if the question were reframed by asking why the highest Court of their State should decline to give effect to its own considered views merely because the highest intermediate appellate court in another (albeit) Australian jurisdiction had reached a different view a short time earlier.
The same hypothetical public may well ponder why judges on the same appellate Court come to disagree sharply on any manner of questions. The simple fact is that disagreement is an inevitable and healthy aspect of a system which engages human actors and requires individual judges to explain their true reasons for joining in the order emerging from a proceeding in their Court (see generally M D Kirby, “Judicial Dissent – Common and Civil Law Traditions” (2007) 123 LQR 379).
Each judge is required to work within the confines of the judicial oath and to apply legal principles, including those principles that form part of the law of precedent. I remain to be persuaded that, prior to the decision of Giles JA and Ipp JA in this case, there is a binding principle of precedent that applies in the present situation.
Having regard to the subject matter of the present legislation and the relative strength of my views as to the interpretation of the key provision, I respectfully decline to follow the Victorian decision.
I nevertheless agree with Ipp JA and Giles JA that the appeal should be dismissed even if the section is to be interpreted as I prefer.
GILES and IPP JJA: This is an application for leave to appeal and, if granted, an appeal against an order made by Bell J under s 17(1)(b) of the Crimes (Serious Sex Offenders) Act 2006 (NSW) (see Attorney General for theState of New South Wales v Tillman [2007] NSWSC 605) that the claimant be detained in a correctional centre for one year from the date of her Honour’s judgment (namely, 18 June 2007).
On 18 April 2007, the claimant completed a sentence of ten years’ imprisonment imposed on him following his conviction for a serious sexual offence committed against the nine-year old daughter of his then de facto wife.
On the same date, the claimant was released subject to an interim supervision order. He remained at liberty until 3 May 2007 when the Court of Appeal ordered that he be detained in a correctional centre on an interim basis for a period of 28 days from 3 May 2007 (Attorney Generalfor New South Wales v Tillman [2007] NSWCA 119). On 29 May 2007, the interim detention order was renewed as from 30 May 2007 for a further 28 days, or until further order. On 18 June 2007, Bell J made the order, the subject of the present application for leave to appeal, in terms of which the claimant is to be detained until 17 June 2008.
The Crimes (Serious Sex Offenders) Act commenced on 3 April 2006. In the Second Reading Speech (New South Wales Parliamentary Debates (Hansard), Legislative Council, 30 March 2006 (The Hon Tony Kelly) at 21801), the Minister for Justice is reported as having said (at 21801 to 21802):
“One particular concern, that is dealt with by the scheme, relates to a handful of high-risk, hard-core offenders who have not made any attempt to rehabilitate whilst in prison. These offenders make up a very small percentage of the prison population, yet their behaviour poses a very real threat to the public. These concerns are compounded where the offender never qualifies for parole and is released at the end of their sentence totally unsupervised.
The proposed legislation will address this problem by allowing this small group of high-risk offenders to be placed on extended supervision, or, in only the very worst cases, kept in custody.”
Later, in the Second Reading Speech, the Minister said (at 21803):
“[The Supreme Court] may only make a continuing detention order if it is satisfied, to a high degree of probability, that the offender is likely to commit a further serious sex offence if he or she is not kept in custody.”
Section 3 of the Act provides:
“Objects of Act
The objects of this Act are to provide for the extended supervision and continuing detention of serious sex offenders so as:
(a) to ensure the safety and protection of the community, and
(b) to facilitate the rehabilitation of serious sex offenders.”
It is common ground that the claimant has committed a serious sex offence as defined by the Crimes (Serious Sex Offenders) Act and is a serious sex offender within the meaning of the Act.
As Bell J observed (at [10]):
“The Act empowers the Court on the application of the Attorney General to make an extended supervision order (for which provision is made in Part 2) or a continuing detention order (for which provision is made in Part 3). Each order operates to extend the period of control over a serious sex offender beyond the term of the sentence of imprisonment imposed in consequence of conviction for his or her sex offence.”
Section 17 of the Crimes (Serious Sex Offenders) Act provides:
“Determination of application for continuing detention order
(1)The Supreme Court may determine an application for a continuing detention order:
(a) by making an extended supervision order, or
(b) by making a continuing detention order, or
(c) by dismissing the application.
(2)An extended supervision order may be made if and only if the Supreme Court is satisfied to a high degree of probability that the offender is likely to commit a further serious sex offence if he or she is not kept under supervision.
(3)A continuing detention order may be made if and only if the Supreme Court is satisfied to a high degree of probability that the offender is likely to commit a further serious sex offence if he or she is not kept under supervision and that adequate supervision will not be provided by an extended supervision order.
(4)In determining whether or not to make a continuing detention order or extended supervision order, the Supreme Court must have regard to the following matters in addition to any other matter it considers relevant:
(a) the safety of the community,
(b)the reports received from the psychiatrists appointed under section 15(4) to conduct psychiatric examinations of the offender, and the level of the offender’s participation in any such examination,
(c)the results of any other assessment prepared by a qualified psychiatrist, registered psychologist or registered medical practitioner as to the likelihood of the offender committing a further serious sex offence, the willingness of the offender to participate in any such assessment, and the level of the offender’s participation in any such assessment,
(d)the results of any statistical or other assessment as to the likelihood of persons with histories and characteristics similar to those of the offender committing a further serious sex offence,
(e)any treatment or rehabilitation programs in which the offender has had an opportunity to participate, the willingness of the offender to participate in any such programs, and the level of the offender’s participation in any such programs,
(f)the level of the offender’s compliance with any obligations to which he or she is or has been subject while on release on parole or while subject to an earlier extended supervision order,
(g)the level of the offender’s compliance with any obligations to which he or she is or has been subject under the Child Protection (Offenders Registration) Act 2000 or the Child Protection (Offenders Prohibition Orders) Act 2004,
(h)the offender’s criminal history (including prior convictions and findings of guilt in respect of offences committed in New South Wales or elsewhere), and any pattern of offending behaviour disclosed by that history,
(i)any other information that is available as to the likelihood that the offender will in future commit offences of a sexual nature.
(5)Part 2 applies to an extended supervision order made under this section in the same way as it applies to an extended supervision order made under section 9.”
It follows, by s 17(1), that, in determining an application for a continuing detention order, the Court may make an extended supervision order. The Act sets out a non-exhaustive list of conditions that might be imposed when making a supervision order (s 11). By s 18(1)(b), a continuing detention order expires at the end of the period (not exceeding five years from the day on which it commences) specified in the order. By s 19, the Court may, at any time, vary or revoke a continuing detention order.
An important issue before Bell J was the meaning of “likely” in s 17(3). Her Honour recorded (at [18]) the opponent’s submission as being that, in this section (and in s 17(2)), “likely” is used in what was said to be its ordinary sense and means “substantial – a real and not remote chance regardless of whether it is more or less than 50 per cent”: Boughey v The Queen (1986) 161 CLR 10 at 21 per Mason, Wilson and Deane JJ).
Her Honour considered that the element that the opponent had to prove was the likelihood of a further serious sex offence being committed if the claimant was not kept under supervision, and the standard of proof to be applied was encompassed by the word “likely” in the section.
Bell J construed “likely” in the relevant context as conveying “the sense that it is probable, as distinct from possible, that the [claimant] will commit a further serious sex offence” (at [29]). She did not say, expressly, whether she construed “likely” as meaning “more probable than not”. Her Honour recognised that the effect of the words “high degree of probability”, in s 17(3), required the opponent to establish to a high degree of probability that the defendant was “likely” to commit a further serious sex offence if he was not kept under supervision and that adequate supervision would not be provided by an extended supervision order.
After carrying out a meticulous examination of the relevant evidentiary material, her Honour said (at [165]):
“Taking into account the expert evidence, particularly the opinions of Dr Allnutt and Professor Greenberg and looking at the inferences to be drawn from the [claimant’s] pattern of past offending, upon the standard of proof of high probability, I consider the [opponent] has established that it is likely that the [claimant] will commit a further serious sex offence if he is not kept under supervision.”
Before this Court, in written submissions made on the claimant’s behalf, it was argued that “likely” must be given the meaning, “a high degree of probability”. In oral submissions, Mr Haesler SC, who, together with Ms Beckett, appeared for the claimant, submitted that the meaning of “likely” went “beyond more probable than not”.
It is well established that the word “likely” is capable of different shades of meaning (McClellan CJ at CL, in Attorney General for the State of New South Wales v Winters [2007] NSWSC 1071, gives several examples). Boughey illustrates the point. This case concerned s 157(1) of the Criminal Code Act 1924 (Tas) which provided:
“(1) Culpable homicide is murder if committed –
…
(c)by means of an unlawful act or omission which the offender knew, or ought to have known, to be likely to cause death in the circumstances, although he had no wish to cause death or bodily harm to any person.”
Gibbs CJ (at 14) referred to cases where the High Court had treated the words “likely” and “probable” as synonymous. The cases in question concerned the mental element necessary at common law to constitute the crime of murder. His Honour said (at 14) that the word “likely” in s 157 of the Tasmanian statute meant “probable” and not “possible”. The Chief Justice observed that this was the “natural meaning” of “likely” and went on to say (at 15) that, in directing a jury as to the meaning of the section, a trial judge should not put a gloss on the ordinary words used. In particular, his Honour said that a judge directing a jury as to the effect of the section should avoid the use of the word “chance”, even if qualified by such words as “good”, “substantial” or “real”. He said (at 15) that “[i]t would however be helpful to the jury to explain that a possibility, as distinct from a probability, is not enough”.
As we have mentioned, Mason, Wilson and Deane JJ said (at 21) that the word “likely” was used in the section “with what we apprehend to be its ordinary meaning, namely, to convey the notion of a substantial – a ‘real and not remote’ – chance regardless of whether it is less or more than 50 per cent.”
Brennan J construed “likely” as “probable” (see at 42 et seq). The equation with “probable”, however, may not assist, as “probable” has the same capacity for shades of meaning from more probable than not to a possibility: see Darkan v The Queen (2006) 227 CLR 373 (at 381 to 382, [23] to [27]).
Of course, the Crimes (Serious Sex Offenders) Act is a different statute to that construed in Boughey and the context is different.
There are other cases where courts have construed the word “likely” in a similar but not the same context as the Crimes (Serious Sex Offenders) Act. Of the cases that fall into this category the most relevant is TSL v Secretary tothe Department of Justice (2006) 14 VR 109. Section 11(1) of the Serious Sex Offenders Monitoring Act 2005 (Vic) (“the Monitoring Act”) provides:
“A Court may only make an extended supervision order in respect of an offender if it is satisfied, to a high degree of probability, that the offender is likely to commit a relevant offence if released in the community on completion of the service of any custodial sentence that he or she is serving, or was serving at the time at which the application was made, and not made subject to an extended supervision order.” [emphasis added]
The scheme of the Monitoring Act differs from the Crimes (Serious Sex Offenders) Act. Moreover, in TSL, the Victorian Court of Appeal appears to have had some regard to the Charter of Human Rights and Responsibilities Act 2006 (Vic), noting (at footnote 15, p 113) that, while the Charter was not yet in force, “the nature of our society is a legitimate factor to take into account in construing the legislation”. Nevertheless, the Monitoring Act is sufficiently close to the Crimes (Serious Sex Offenders) Act as to require this Court to follow TSL unless it is satisfied that TSL is clearly wrong.
Callaway AP (with whom Buchanan JA and Coldrey AJA agreed) said (at 112, [9]):
“Because it was concerned with the future, Parliament could not require the court to be satisfied that the offender will commit a relevant offence. All that the Court could be satisfied of is that the offender is likely to do so or that there is a risk that the offender will do so.”
Callaway AP said (at 113, [9] and [10]) that construing “likely” as connoting “a high degree of probability” (in the particular context) was the preferred meaning of “likely” in the section. Significantly, however, his Honour went on to say (at 113, [11]):
“I desire to guard against misunderstanding. To say that ‘likely’ is used in that sense does not mean that the secretary must prove that it is more likely than not that a person will commit a relevant offence. Likelihood, in the ordinary sense of that word identified in Boughey v R, includes a low degree of probability. What the legislature requires the court to be satisfied of is a high degree of probability. There is no reason to think that it must be more than 50%. With experience, it may be possible to be more precise, but we should always return to the words the legislature has chosen and the context in which they appear.”
Thus, in TSL, the Victorian Court of Appeal construed “likely” to mean “probable” in the sense of “a high degree of probability”, but not necessarily involving a degree of probability of more than 50 per cent.
In Attorney General for the State of New South Wales vGallagher [2006] NSWSC 340, McClellan CJ at CL made an interim detention order under s 16 of the Crimes (Serious Sex Offenders) Act. In the course of his Honour’s decision he commented (at [34]) in relation to s 17(2):
“The word ‘likely’ presently appears to me to be used in its ordinary meaning which, as the High Court pointed out in Boughey v The Queen (1986) 161 CLR [10 at 21] … conveys the notion of [a] ‘substantial – a real and not remote – chance regardless of whether it is less or more than 50 per cent’”.
His Honour, nevertheless, expressed the reservation (at [35]) that it would be necessary to give careful consideration to the context in which the word “likely” is used “upon the final determination of this matter”.
The Chief Judge returned to the issue again in Winters. In commenting on the judgment of Bell J in [2007] NSWSC 605, his Honour said (at [29]):
“Her Honour was of the view that likely meant ‘probable’ as distinct from ‘possible’. Although she did not state whether she accepted that it meant ‘more probable than not’, I infer that this was the meaning which her Honour accepted to be appropriate.”
His Honour undertook the consideration foreshadowed in Gallagher, and concluded (at [50]):
“[I]n my judgment, in the Act, ‘likely’ means ‘more likely than not’ and it is that test which must be applied in ss 9 and 17. For relevant purposes the court must be satisfied to a high degree of probability that it is more likely than not that the offender will commit a further serious sex offence. It follows that the defendant’s argument that the chance of reoffending must be significantly higher than 50% must also be rejected.”
The Chief Judge said of TSL (at [29]) that “leaving aside matters of comity”, he would not favour the construction adopted in that case.
We now turn to the exercise of construction required by this Court.
In our opinion, the first step is to recognise the legislature was dealing with a future state of affairs, the existence of which could not be known with certainty (see TSL at 112, [9]).
Next, due regard must be had to the fact that s 17(3) provides for the continued detention of a person who has served the sentence that was imposed on him or her, and the legislation involves the curtailment of the liberty of an individual. In Re Bolton; Ex parte Beane (1987) 162 CLR 514, Brennan J said (at 523):
“The law of this country is very jealous of any infringement of personal liberty … and a statute or statutory instrument which purports to impair a right to personal liberty is interpreted, if possible, so as to respect that right.”
See also Plaintiff S157/2002 v The Commonwealth of Australia (2003) 211 CLR 476 where Gleeson CJ said (at 492, [30]):
“[C]ourts do not impute to the legislature an intention to abrogate or curtail fundamental rights or freedoms unless such an intention is clearly manifested by unmistakable and unambiguous language. General words will rarely be sufficient for that purpose. What courts will look for is a clear indication that the legislature has directed its attention to the rights or freedoms in question, and has consciously decided upon abrogation or curtailment”.
Callaway AP took account of the two factors we have mentioned when saying in TSL (at 113, [10]):
“It is understandable that Parliamentary counsel would have chosen the word ‘likely’ in relation to a future state of affairs but almost inconceivable that Parliament would have intended that word to bear its ordinary meaning. All too many offenders are likely, in that sense, to commit a relevant offence. A person subject to an extended supervision order is a prisoner in all but name. The threshold would be far too low, in a free society, if a Court had a discretion to make an extended supervision order simply because it was satisfied that there was ‘a substantial – ‘real and not remote’ – chance’ of his or her re-offending. That is why the word ‘likely’ in s 11(1) is used in a sense of a high degree of probability.”
His Honour’s observations apply, all the more so, to a continuing detention order.
The canon of construction exemplified by the remarks of Brennan J in Re Bolton; Ex parte Beane, and Gleeson CJ in Plaintiff S157/2002 v Commonwealth, respectively, supports construing “likely” as denoting a relatively high degree of probability. On the other hand, there are a number of indicia in the Crimes (Serious Sex Offenders) Act that are contrary to the proposition that “likely” should be construed to mean “beyond more probable than not” (as the claimant argued).
Firstly, s 17(3) uses the expressions “high degree of probability” and “likely” in close proximity. There is, therefore, a strong inference that Parliament intended to distinguish between the meanings conveyed by the two expressions. Applying their natural meanings, “high degree of probability” denotes a higher degree of probability than “likely”. Linguistically, one would more readily attribute a meaning of “beyond more probable than not” to a “high degree of probability” than to “likely”.
Secondly, giving “likely” the meaning of “high degree of probability” (in ss 17(2) and (3)) creates an oddity of language. One would be left with a section reading something like “a continuing detention order may be made if and only if the Supreme Court is satisfied to a high degree of probability that it is highly probable that the offender is likely to commit a further serious sex offence if he or she is not kept under supervision”. It would have been far easier for Parliament, had such a meaning been intended, to provide, “an extended supervision order may be made if and only if the Supreme Court is satisfied that there is a high probability of the offender committing a further serious sex offence if he or she is not kept under supervision”. The repetition of the notion of “a high degree of probability” involved in the construction for which the claimant contends, militates – linguistically - against that construction.
Thirdly, the repeated use of the word “likelihood” in ss 17(4)(c), 17(4)(d) and 17(4)(i) (and the omission, in these sections, of the expression “high probability”) supports the inference that the notion of “likely” and “likelihood” is intended to have a meaning different to “a high degree of probability”.
In our view, there is much to be said for the view expressed by McClellan CJ at CL in Winters, namely, that “likely” is used in the section as meaning more probable than not. On the other hand, it is a respectable view that the legislature recognised the infringement of personal liberty by requiring satisfaction to a high degree of probability, so that it would be wrong to impose in the shade of meaning adopted for “likely” a further strict requirement of likelihood in the sense of more probable than not. There is ambiguity in the word “likely”, and if unconstrained we would tend towards his Honour’s view. Nevertheless, the view expressed in TSL is reasonably open and we are not persuaded that it is clearly wrong. In the light of what has emanated from the High Court in regard to the respect that an intermediate appellate court of one Australian jurisdiction should give to a decision of an intermediate appellate court of another Australian jurisdiction on issues that are substantially the same, we would follow and adopt the approach of Callaway AP in TSL.
Accordingly, we would hold that the word “likely” in ss 17(2) and (3) denotes a degree of probability at the upper end of the scale, but not necessarily exceeding 50 per cent. The remarks Bell J made when articulating the test she intended to apply are to be understood in this sense. Her Honour said (at [29]):
“I have approached the determination in this case upon the basis that ‘likely’ in this context conveys the sense that it is probable, as distinct from possible, that the defendant will commit a further serious sex offence.”
Accordingly, for the reasons stated, we differ, with respect, from what McClellan CJ at CL said in Winters on this issue. As well, we do not understand Bell J to have meant “probable” as meaning “more probable than not”. In our opinion, nothing her Honour said suggests this inference. For the reasons we have stated, we would hold that, in adopting this more limited approach, her Honour was correct. For the reasons we have explained, while “likely” means “probable”, it does not mean “more probable than not”.
Bell J said (at [30]):
“The [opponent] must establish to a high degree of probability that (i) the [claimant] is likely to commit a further serious sex offence if he is not kept under supervision, and (ii) that adequate supervision will not be provided by an extended supervision order. If the [opponent] establishes both the Court may make a detention order. In the event the [opponent] establishes (i) but does not establish (ii) the Court may make an extended supervision order. The Court retains a discretion not to make either order.”
In our opinion, with the understanding of likelihood adopted by her Honour, this approach discloses no error.
The difference between likelihood in the sense of a high probability but not necessarily more probable than not, and likelihood as something more probable than not, may not be great. Expressed as percentages, which is incorrect because it suggests a mathematical precision which is unattainable and is an unhelpful approach, transition from 49 per cent to 51 per cent is not the key to application of ss 17(2) and (3).
We therefore come to the conclusion that the claimant’s arguments as to the meaning of “likely” in s 17(3) cannot be accepted.
We would add that, even on the claimant’s construction of s 17(3), we would not interfere with her Honour’s decision. Her Honour’s careful and thorough examination of all the facts demonstrates that, even on the test for which the claimant contends, the degree of satisfaction required by s 17(3) was established.
We would grant the application for leave to appeal but dismiss the appeal.
Since writing the foregoing we have had the benefit of reading in draft the reasons of Mason P. We remain of the view that on comity grounds the approach of Callaway AP in TSL should be adopted, but more full explanation of that view is called for.
While a judge’s obligation to the law and to the judge’s own conscience requires the exercise of an independent judgment, that is done within established principles of binding precedent and deference to prior decisions. The deference may be afforded to decisions of the judge’s own court or to decisions of another court. While it is often said that a prior non-binding decision is followed out of comity, more than courtesy is involved. Both precedent and deference serve the important purpose of certainty and predictability in the law. Business and personal affairs can be conducted with confidence in reasonable legal stability, and even if there is not express or implicit reliance on existing expositions of the law in the conduct of affairs the same confidence can underpin advice in the event of civil dispute or a criminal charge. Certainty and predictability are a vital part of the rule of law.
The principles of precedent and deference to prior decisions have developed over recent years. In their development there can be seen increasing regard to an Australian national system of law.
Reflecting Australia’s legal heritage, in earlier times decisions of the English Court of Appeal and the House of Lords were regarded as binding or at least to be followed as a general rule: eg Public Transport Commission of New South Wales v J Murray-More (NSW) Pty Ltd (1975) 132 CLR 336 at 341 (Barwick CJ), 349 (Gibbs J); Viro v The Queen (1978) 141 CLR 88 at 121 (Gibbs J). In Parker v The Queen (1963) 111 CLR 610, the High Court departed from its practice of following the House of Lords. In Australian Consolidated Press Ltd v Uren [1969] 1 AC 590, the Privy Council recognised that the Australian common law might diverge from that of England. Appeals to the Privy Council were not possible from the High Court by 1975 and from State Courts by 1986. In Cook v Cook (1986) 162 CLR 376 at 390, it was said that statements that a State Supreme Court, including on appeal, should as a general rule follow decisions of the English Court of Appeal were no longer binding upon Australian courts. There was a possible qualification as to decisions of the House of Lords given during the period in which appeals lay to the Privy Council, but the autonomy of Australian courts in expressing a national law was enhanced. There have been significant departures from the common law of England: see Toohey, “Towards an Australian Common Law” (1990) 6 Australian Bar Review 185 at 193 to 195.
Within Australia, it has been accepted that Commonwealth legislation should receive consistent interpretation. In R v Abbrederis [1981] 1 NSWLR 530, Street CJ followed a decision of the Full Court of the Supreme Court of Victoria construing Commonwealth legislation notwithstanding that he disagreed with the construction, saying (at 542):
“As a matter of precedent this Court is not, of course, bound by the decision of the Full Court of Victoria. But I have not the slightest doubt that, where a Commonwealth statute has been construed by the ultimate appellate court within any State or Territory, that construction should, as a matter of ordinary practice, be accepted and applied by the courts of other States and Territories so long as it is permitted to stand unchanged either by the court of origin or by the High Court. The risk of differing interpretations amongst the States is thus negated and, in practical terms, a uniform application of Commonwealth laws throughout Australia is assured.”
Courts of other states have taken the same stance, see for example the trilogy of R v Gardiner [1981] Qd R 394 (Court of Criminal Appeal (Queensland)), Zibillari v The Queen [1981] WAR 40 (Court of Criminal Appeal (Western Australia)) and R v Parsons [1983] 2 VR 499 (Full Court of the Supreme Court of Victoria). The Full Federal Court has been regarded as in the same position as a State intermediate appellate court: R v Yates (1991) 102 ALR 673.
This has been extended to uniform national legislation. In Australian Securities Commission v Marlborough Gold Mines Ltd (1993) 177 CLR 485, the High Court said (at 492):
“Although the considerations applying are somewhat different from those applying in the case of Commonwealth legislation, uniformity of decision in the interpretation of uniform national legislation such as [the Corporations Law] is a sufficiently important consideration to require that an intermediate appellate court – and all the more so a single judge – should not depart from an interpretation placed on such legislation by another Australian intermediate appellate court unless convinced that that interpretation is plainly wrong.”
In Re J & E Holdings Pty Ltd and the Corporations Law (1995) 36 NSWLR 541, Sheller JA, with whom Priestley and Powell JJA agreed, said of the interpretation placed by the Appellate Division of the Victorian Supreme Court on a provision of the Corporations Law (at 551):
“This is the latest decision by an intermediate appellate court on the point. Their Honours carefully reviewed the arguments and concluded that the majority decision of the Queensland Court of Appeal was plainly wrong. Even if I retained some residual doubt about the Victorian Court's conclusion I think we should follow it. To do otherwise would seem to me to be giving but lip service to what the High Court has said. Certainty in the law, in my opinion, requires that only in an extreme case would an intermediate appellate court or a judge of first instance not follow the latest decision by an intermediate appellate court if, in that latest decision, the arguments have been fully reviewed and a conclusion reached that an earlier decision of another intermediate appellate court was plainly wrong.”
It has been emphatically stated that:
“There is but one common law in Australia … In contrast to the position in the United States, the common law as it exists throughout the Australian States and Territories is not fragmented into different systems of jurisprudence, possessing different content and subject to different authoritative interpretations.”
Lange v Australian Broadcasting Corporation (1997) 189 CLR 520 at 563; see also Esso Australia Resources Ltd v Commissioner of Taxation (1999) 201 CLR 49. In Farah Constructions Pty Ltd v Say-Dee Pty Ltd (2007) 81 ALJR 1107, the High Court said (at 1140, [135]):
“Intermediate appellate courts and trial judges in Australia should not depart from decisions in intermediate appellate courts in another jurisdiction on the interpretation of Commonwealth legislation or uniform national legislation unless they are convinced that the interpretation is plainly wrong. Since there is a common law of Australia rather than of each Australian jurisdiction, the same principle applies in relation to non-statutory law.”
Commonwealth legislation, uniform national legislation and the common law have obvious claims to national certainty and predictability. The first and third are truly nation-wide, the secondly is effectively nation-wide, and there should be consistent decision-making throughout Australia notwithstanding the existence of separate legal jurisdictions. Perpetuation of egregious error is countered by departure from the prior decision if persuaded that it is plainly wrong, but there is for the most part social and economic unity within Australia calling for comity (in the sense described above) between the appellate courts of the separate jurisdictions.
Although the High Court has not yet spoken, there is support in the cases for similarly following decisions of other State and Territory courts at the same judicial level on the interpretation of substantially similar State or Territory legislation. Mason P has referred to a number of the cases.
In Fernando v Commissioner of Police (1995) 36 NSWLR 567, there was a decision of the Full Court of the Supreme Court of South Australia construing a substantially similar statutory provision. Priestley JA recognised (at 584) “the rule of comity pursuant to which [this Court] usually follows the decisions of courts of like position in other Australian jurisdictions on similar points”. Powell JA noted (at 593) the high desirability of conformity of decisions not only in the construction of Commonwealth legislation and uniform national legislation but also in the case of “legislative provisions which are identical in terms, or substantially so”. Both their Honours did not follow the South Australian decision: Priestley JA considered that there were “strong reasons” for departing from it and Powell JA considered that it was clearly wrong.
Clarke JA followed the decision, saying at 589 to 590 that he considered it settled that this Court should follow the decision of another Australian intermediate appellate court on a statutory provision substantially identical with the one under consideration unless convinced that the early decision was clearly wrong or that considerations of justice required the Court to decline to apply it; his Honour continued:
“This is, to my mind, a most important principle. It promotes greater uniformity and certainty in the law and avoids the plainly unsatisfactory situation which would arise if particular statutory words are regarded as having different meanings in different parts of the country.
In past years, the general acceptance by Australian courts of the principle of law developed and explained in the United Kingdom had led to a degree of uniformity. But the position has changed and, in some respects, quite dramatically. There now is increasing divergence between English law and the body of Australian law, which in the final analysis, is the law developed in the High Court of Australia. It is that Court which acts as the unifying force of Australian law but where it has not laid down binding principles intermediate appellate courts are free, within the limits of comity, to develop their own principles. In the absence of strict adherence to the principle under discussion, it would be idle to speak of Australian law except where the High Court had intervened. It would be more appropriate to speak of the law of a State, Territory or the Commonwealth. But if the courts of the States, Territories and Commonwealth are engaged, in general, in developing and applying principles of Australian law, as I believe they are, then a control is necessary to prevent the unbridled development of divergent principles.
In these circumstances, it is far preferable that an intermediate appellate court apply principles expressed in an earlier decision of another intermediate appellate court, although, perhaps, doubting its correctness, leaving it to the High Court to lay down the binding rule. In this way the law remains constant and certain throughout the country until, as may occur, the High Court expresses a different view which will, in essence, also bind courts all over Australia.
In an important respect the position is different from the one which arises when an appellate court, be it of a State or Territory reconsiders an earlier decision of its own and declines to apply it. In that instance the law as laid down in the latter case is taken to be the law of the State or Territory. Where, however, one State appellate court declines to follow a decision of another appellate court in Australia the consequence is, or may be, that there is an immediate inconsistency between the respective States or Territories and, of course, trial courts will be bound by the decision of the appellate court in their State or Territory. For this reason even greater caution should be exercised in deciding not to follow the existing authority of another State or Territory.”
In R v NZ (2005) 63 NSWLR 628, Howie and Johnson JJ, with whom Wood CJ at CL and Hunt AJA relevantly agreed, referred (at 666, [160]) to the “rule of comity referred to in Fernando” as having application “where the court is considering a decision of another Australian intermediate appellate court that has interpreted an identical or substantially similar provision in a State statute”. Their Honours said (at 668, [166]) that it was not an inflexible and universal formula, and that the weight and respect attached to the decision of an intermediate appellate court concerning similar legislative provisions might be reduced where it was not the proper construction of the legislation that was under consideration but rather issues of practice and procedure involving the operation of the statutory provisions in their local context. That is not the present case.
Again, courts of other States have taken a like stance, see Sullivan v The Queen (1987) 47 NTR 31 cited by Mason P and R v Lorkin (1995) 15 WAR 499 in which the Full Court of the Supreme Court of Western Australia, in following the Queensland Court of Appeal on identical legislation, said that the decision was not binding but “of persuasive authority”.
We respectfully adopt Clarke JA’s explanation for the importance of comity between intermediate appellate courts in the construction of substantially similar State or Territory legislation. The place now occupied by the Federal Court in the Australian legal system adds to its unity, and to the desirability of consistency in interpretation of substantially similar non-national legislation. The public would be entitled to ask why the same words of a statute have different meanings in different States or Territories, in the present case different meanings north and south of the Murray River. Being unpersuaded that TSL is plainly wrong, we consider that it should be followed.
**********
LAST UPDATED: 26 November 2007
87
28
6