Darkan v The Queen - [2006] HCA 34

HIGH COURT OF AUSTRALIA
GLEESON CJ
GUMMOW, KIRBY, HEYDON AND CRENNAN JJ
Matter No B87/2005
HOWARD RODNEY DARKAN  APPELLANT
AND
THE QUEEN  RESPONDENT
Matter No B88/2005
GWENDOLINE CECILY DEEMAL-HALL  APPELLANT
AND
THE QUEEN  RESPONDENT
Matter No B89/2005
MARLOW PHILLIP ANDREW McIVOR  APPELLANT
AND
THE QUEEN  RESPONDENT
Darkan v The Queen
Deemal-Hall v The Queen
McIvor v The Queen
[2006] HCA 34
22 June 2006
B87/2005, B88/2005 & B89/2005
ORDER
In each matter, the appeal is dismissed.

On appeal from the Supreme Court of Queensland
Representation
A J Rafter SC for the appellant in B87/2005 (instructed by Legal Aid Queensland)
P J Callaghan SC with A W Moynihan for the appellant in B88/2005 (instructed by Legal Aid Queensland)
M J Byrne QC for the appellant in B89/2005 (instructed by Legal Aid Queensland)
M J Copley for the respondent in all matters (instructed by Director of Public Prosecutions (Queensland))
Notice: This copy of the Court's Reasons for Judgment is subject to formal revision prior to publication in the Commonwealth Law Reports.

CATCHWORDS
Darkan v The Queen
Deemal-Hall v The Queen
McIvor v The Queen
Criminal Law – Common intention to prosecute unlawful purpose – Aiding the commission of an offence – Counselling the commission of an offence – Assault causing death – Appellants tried together and convicted of murder – Where Crown case relied on ss 7, 8 and 9 of the Criminal Code (Q) – Jury directed by trial judge that for the purpose of ss 8 and 9 of the Criminal Code (Q) "a probable consequence" was one which was "a real possibility or a substantial chance or a real chance" – Meaning of "a probable consequence" – Whether "probable" connotes something more than "possibility" or "real possibility" or "real chance" – Whether jury was misdirected.
Criminal Law – Application of the proviso under s 668E(1A) of the Criminal Code (Q) – Whether "substantial miscarriage of justice" has actually occurred – Whether trial fundamentally flawed – Relevance of unknown mode of jury reasoning – Relevance of the fact that the misdirection concerned the elements of the offences charged.
Appeal – Court of criminal appeal – Proviso not considered by court of criminal appeal – Whether High Court should consider proviso in the circumstances – Role of court of criminal appeal – Role of High Court – Limitations on High Court considering the proviso without scrutiny of the record of evidence by court of criminal appeal – Defects of a trial on the record.
Words and phrases – "a probable consequence", "substantial miscarriage of justice".
Criminal Code (Q), ss 8, 9, 668E(1A).

GLEESON CJ, GUMMOW, HEYDON AND CRENNAN JJ. These three appeals were argued together. The appellant in the first appeal is Howard Rodney Darkan ("the first appellant"), in the second Gwendoline Cecily Deemal-Hall ("the second appellant") and in the third Marlow Phillip Andrew McIvor ("the third appellant"). The appellants were jointly tried at Cairns before a jury in the Supreme Court of Queensland (Jones J). They were charged with murdering Kalman John Toth ("the deceased") on 13 January 2003. They were each convicted and sentenced to life imprisonment. Their appeals to the Court of Appeal against conviction were dismissed[1]. The appellants appeal against those orders on a single ground relating to the directions which the trial judge gave the jury on the meaning of the expression "a probable consequence" as used in ss 8 and 9 of the Criminal Code (Q) ("the Code"). In the case of all three appellants, the prosecution relied on s 8. In the case of the second appellant, it relied also on s 9. The appellants contend that the trial judge erred in telling the jury that "a probable consequence" was one which was "a real possibility or a substantial cause or a real chance".
[1]R v Deemal-Hall,Darkan & McIvor [2005] QCA 206.
That contention is sound. However, each appeal should be dismissed because the proviso should be applied – that is, because "no substantial miscarriage of justice has actually occurred" within the meaning of those words in s 668E(1A) of the Code.
The factual circumstances
The appellants did not give evidence at the trial. Each appellant was interviewed by police officers, but the answers of any particular appellant were only admissible as against that appellant, and the admissions made by the second appellant were far less extensive than those made by the other two. For that reason it is necessary to be careful in assessing the strength of the case against each appellant in relation to s 668E(1A). However, for present purposes, the primary evidence admissible against all the appellants, largely given by a witness named Bowen, revealed the prosecution case to be as follows.
The second appellant had been the de facto wife of the deceased. She had a grievance against him and wanted to cause him some harm. The deceased was aged 58, but he was a large man, and the second appellant saw it as necessary to recruit the services of no fewer than three men to cause him harm – the first appellant, aged 30, the third appellant, aged 23, and Bowen, aged 22. Indeed, she attempted to secure as well the services of a fourth man named Michael James Cobus. The evidence about the nature of the harm she requested will be examined below. For their services, each of the three men recruited was to receive $50, and more the following week.
The second appellant took the first appellant, the third appellant and Bowen in a van to Bicentennial Lakes Park, Mareeba, on the evening of 13 January 2003. She dropped them there and later returned with the deceased. The deceased got out of the van, but later went back into it and locked it. The second appellant persuaded him to let her enter it, and to drive it to a shed area. They then alighted and sat down with the first appellant, the third appellant and Bowen. The first appellant began to taunt the deceased and punched him in the face. The deceased stood up and began defending himself, and a fist fight broke out. It continued for some time. The third appellant then hit the deceased on the back of the neck with a pickaxe handle. The deceased fell to the ground and covered his face up as the first appellant and the third appellant, who (at least according to Bowen) were wearing steel-capped boots, began kicking him. They continued to kick him for a good while. Watched by the second appellant, the first appellant then took the pickaxe handle from the third appellant, and hit the deceased for a couple of minutes around the ankles and knees before making his way up the body and hitting the deceased in the ribs. The deceased was crying for help. Bowen took the pickaxe handle from the first appellant and gave it to the third appellant. He told him to get rid of it and not to give it back to the first appellant. However, the first appellant took the pickaxe handle away from the third appellant and began hitting the deceased in the head as if "he really meant it this time".
The appellants and Bowen then panicked and left the scene. Before he left, Bowen examined the deceased's face and found him to be "messed up pretty bad". The deceased asked Bowen to get help, and Bowen told the second appellant of this request before she dropped him at his aunt's house and gave him his $50.
The deceased's body was discovered at 6am the following day. A pathologist who conducted a post mortem examination found that, apart from severe bruising of many parts of the deceased's body, there was fracturing of the upper and lower jaw and of many facial bones, that there were many facial lacerations, and that the cause of death was aspiration of blood due to severe facial trauma. The pathologist thought that the injuries could have been caused by the pickaxe handle, if severe force were used.
What had the second appellant said to trigger the death of the deceased in this way? On Bowen's account in chief of a conversation between himself, the first appellant and the second appellant, the first appellant said that the second appellant wanted someone to "fix him up" (ie the deceased), and the second appellant said she would pay whoever would "fix it up". Bowen also said in chief that while the second appellant was driving the three men in the van to the park, she said she "just wanted someone to get into him". He repeated that evidence in cross-examination. Counsel for the second appellant cross-examined Bowen as follows:
"Mr Bowen, whatever the exact words being used in the conversation were, whatever they might have been, is it the case that what was discussed was a plan to give this fellow a touch-up, is that correct?-- Yes.
And nothing more than a touch-up?-- Yes."
Bowen repeated that evidence twice. He also said that there had been no discussion of a plan to kick the deceased or use a weapon on him.
The legislation
Section 291 of the Code provides that it is unlawful to kill any person unless the killing is authorised or justified or excused by law. Section 293 provides that any person who causes the death of another, directly or indirectly, by any means whatever, is deemed to have killed that other person. Section 300 provides that any person who unlawfully kills another is guilty of murder or manslaughter according to the circumstances of the case. The following provisions of the definition of "murder" in s 302 are relevant.
"(1)Except as hereinafter set forth, a person who unlawfully kills another under any of the following circumstances, that is to say –
(a)if the offender intends to cause the death of the person killed or that of some other person or if the offender intends to do to the person killed or to some other person some grievous bodily harm;
(b)if death is caused by means of an act done in the prosecution of an unlawful purpose, which act is of such a nature as to be likely to endanger human life;
...
is guilty of murder."
Chapter 2 of the Code is entitled "Parties to offences". It comprises ss 7-10A.
The heading to s 7 is "Principal offenders". Section 7(1) provides in part:
"When an offence is committed, each of the following persons is deemed to have taken part in committing the offence and to be guilty of the offence, and may be charged with actually committing it, that is to say –
(a)every person who actually does the act or makes the omission which constitutes the offence;
...
(c)every person who aids another person in committing the offence;
(d)any person who counsels or procures any other person to commit the offence."
The heading to s 8 is "Offences committed in prosecution of common purpose". It provides:
"When 2 or more persons form a common intention to prosecute an unlawful purpose in conjunction with one another, and in the prosecution of such purpose an offence is committed of such a nature that its commission was a probable consequence of the prosecution of such purpose, each of them is deemed to have committed the offence."
The heading to s 9 is "Mode of execution immaterial". Section 9(1) provides:
"When a person counsels another to commit an offence, and an offence is actually committed after such counsel by the person to whom it is given, it is immaterial whether the offence actually committed is the same as that counselled or a different one, or whether the offence is committed in the way counselled, or in a different way, provided in either case that the facts constituting the offence actually committed are a probable consequence of carrying out the counsel."
The heading to s 10A is "Interpretation of ch 2". Section 10A(2) provides:
"Under section 8, a person's criminal responsibility extends to any offence that, on the evidence admissible against him or her, is a probable consequence of the prosecution of a common intention to prosecute an unlawful purpose, regardless of what offence is proved against any other party to the common intention."
The prosecution case as put to the jury
As recorded in the trial judge's summing up, the prosecution's case was that either the first appellant or the third appellant or both of them delivered the blows that caused the death of the deceased. It put its case against the first appellant and the third appellant in three ways. The first was that each killed the deceased intending to do him grievous bodily harm: s 302(1)(a) and s 7(1)(a). The second was that each aided the other, or did or omitted to do an act for the purpose of enabling or aiding the other, to kill the deceased knowing that the other intended to do the deceased grievous bodily harm: s 302(1)(a) and s 7(1)(c). The third – and this case was put against the second appellant as well – depended on s 8 of the Code, and will be described as the s 8 allegation. The s 8 allegation was that the appellants formed a common intention to prosecute an unlawful purpose; that the death was caused by an act done in the prosecution of that purpose which was of such a nature as to be likely to endanger human life; and that the act causing death was a probable consequence of the prosecution of the unlawful purpose: s 302(1)(b) and s 8.
As recorded in the trial judge's summing up, the prosecution put its case against the second appellant in a manner additional to the s 8 allegation. It will be described below as the s 9 allegation. It was that she counselled or procured the first appellant and the third appellant to assault the deceased; that they engaged in conduct – administering blows to the deceased with intent to do him grievous bodily harm – which constituted murder as defined in s 302(1)(a); and that the delivery of blows with that intent was a probable consequence of the counselling or procuring: s 7(1)(d) and s 9.
The s 8 allegation, made against all three appellants, in part depended on the meaning of "a probable consequence" in s 8. The s 9 allegation, made against the second appellant, depended in part on the meaning of "a probable consequence" in s 9. The jury was given copies of ss 7, 8 and 9.
The passage complained of
The only complaint which the appellants made to this Court about the summing up concerns a single sentence in a summing up, the transcript of which ran to over 40 pages. The sentence appeared after the trial judge had explained the way in which the prosecution put its s 9 allegation against the second appellant, and just before the trial judge explained the way the prosecution put its s 8 allegation against all three appellants. The sentence concerned the meaning of the expression "a probable consequence" as it appears in s 9. The sentence was:
"Now when I speak of probable consequences, it means that it's a real possibility or a substantial cause or a real chance that that event would happen."
Three preliminary points may be made about this direction.
First, even if the trial judge actually said "substantial cause" rather than "substantial chance" while intending to say "substantial chance", which is open to question, no appellant submitted that anything turned on that slip. It is accordingly safe to decide the appeals on the assumption that instead of "cause" the trial judge said "chance".
Secondly, although this direction was given in relation to s 9, and thus only in relation to the second appellant, in argument before this Court the prosecution did not deny that the jury would or could have taken the direction to apply also to s 8, and hence to the case against all three appellants.
Thirdly, none of the counsel appearing for the appellants at the trial asked that the direction be corrected and given in different terms. There would have been no point in asking for this, since the trial judge was bound by the decision of the Queensland Court of Appeal in R v Hind and Harwood that the expression "a probable consequence" in s 8 was satisfied if the consequence was a real or substantial possibility or chance[2]. The Queensland Court of Appeal in the present case held that R v Hind and Harwood reflected a "settled position" on the interpretation of s 8[3]. That proposition is reflected in the Bench Book ordinarily used in Queensland as the basis for formulating directions[4] and it is assumed by the profession in Queensland to be correct[5]. However, the direction was not mandatory, and no counsel asked either that it not be given or that it be withdrawn.
[2]R v Hind and Harwood (1995) 80 A Crim R 105 at 116-117 per Fitzgerald P and 141-142 per Pincus JA. In fact that decision was a case in which the issue turned not on jury direction but on whether the verdict was reasonable. On a strict view it might be said that the case is not an authority on jury direction. But underlying both the issue of jury direction and the issue of reasonableness of verdict is a common question – what does "a probable consequence" mean?
[3]R v Deemal-Hall,Darkan & McIvor [2005] QCA 206 at [55] per Keane JA, Williams JA and Muir J concurring.
[4]At 71.6 and 71.11.
[5]For example, R v Chan [2001] 2 Qd R 662 at 663 [3]; R v Jeffrey [2003] 2 Qd R 306 at 317.
Difficulties in the word "probable"
There are two preliminary points to be made about the use of the word "probable" in expounding and applying rules of law.
The first is that the application of legal tests that turn on questions of probability will vary with the context in which the question is asked. That is, "probability" can denote a variety of degrees of confidence. Probabilities can be of different degrees of strength. This has been recognised in relation to the construction of criminal statutes[6]. It has been recognised outside that field as well. For example, Kitto, Taylor, Menzies and Owen JJ said in Beecham Group Ltd v Bristol Laboratories Pty Ltd[7] that the first question to be addressed in dealing with applications for interlocutory injunctions in patent cases is:
"whether the plaintiff has made out a prima facie case, in the sense that if the evidence remains as it is there is a probability that at the trial of the action the plaintiff will be held entitled to relief ... How strong the probability needs to be depends, no doubt, upon the nature of the rights he asserts and the practical consequences likely to flow from the order he seeks. Thus, if merely pecuniary interests are involved, 'some' probability of success is enough".
They quoted the words of Roper CJ in Eq in Linfield Linen Pty Ltd v Nejain[8]:
"[T]his being an application for an interlocutory injunction I look at the facts simply to ascertain whether the plaintiff has established a fair prima-facie case and a fair probability of being able to succeed in that case at the hearing."
It is also relevant to note an observation made by Kitto J in the course of argument in the Beecham case[9]:
"When it is said that the plaintiff must show a probability of success, that does not mean that he must show that it is more probable than not that he will succeed. It is enough that he show a sufficient likelihood of success to justify in the circumstances the preservation of property."
[6]Boughey v The Queen (1986) 161 CLR 10 at 20 per Mason, Wilson and Deane JJ.
[7](1968) 118 CLR 618 at 622. The force of what was said in this case is not diminished by the fact that the test now preferred in relation to the grant of interlocutory injunctions is different.
[8](1951) 51 SR (NSW) 280 at 281.
[9](1968) 118 CLR 618 at 620.
The second preliminary point is that whatever precise meaning the word "probable" bears in a particular context, it is usually used to establish a contrast to what is "possible". Thus the Concise Oxford English Dictionary defines "possible" as that which "may exist or happen, but that is not certain or probable"[10].
[10]11th ed (2004).
In Overseas Tankship (UK) Ltd v The Miller Steamship Co Pty[11] Lord Reid said of the word "probable" as used in judgments:
"It is used with various shades of meaning. Sometimes it appears to mean more probable than not, sometimes it appears to include events likely but not very likely to occur, sometimes it has a still wider meaning and refers to events the chance of which is anything more than a bare possibility".
The New Zealand Court of Appeal said that those shades of meaning could be found in ordinary popular speech as well as in judgments[12]:
"The two most common meanings are 'more probable than not' and what Lord Reid described as 'likely but not very likely'. We prefer, for present purposes, to say that a probable event, in this second sense of the word, means an event that could well happen. These two most common meanings are both descriptive of a stronger prospect of the occurrence of an event than is conveyed by the word 'possible'. We see no justification for reading 'probable consequence' in s 66(2)[13] as 'possible consequence'. On the other hand we do not think that it can be said that 'more probable than not' is clearly the primary meaning of the word. It is of course used a great deal in that sense, but so much depends upon the context in which and the purpose for which it is used in any particular case. Our inquiry must therefore be to ascertain the meaning of 'probable' which will best ensure the attainment of the objects of s 66(2)."
[11][1967] 1 AC 617 at 634-635.
[12]R v Gush [1980] 2 NZLR 92 at 94 per Richmond P, Richardson and O'Regan JJ (emphasis in original).
[13]Crimes Act 1961 (NZ), s 66(2), the equivalent to s 8 of the Code.

In this case the possible meanings of "probable" which were referred to were, in descending order of likelihood:
(a) more probable than not;
(b) a probability of less than 50/50, but more than a substantial or real and not remote possibility;
(c) a substantial or real and not remote possibility;
(d) a possibility which is "bare" in the sense that it is less than a substantial or real and not remote possibility.
The parties agreed that (a) and (d) were incorrect. It was also agreed that (c) was wrong. The appellants contended that (b) was correct. The respondent contended that if (b) was correct, the trial judge's direction did not diverge from it.
Before the arguments advanced by the appellants are considered, it is desirable to examine the history of ss 8 and 9 of the Code; the reasoning advanced in the Queensland authorities in support of the direction impugned in this case; the points of construction of ss 8 and 9 on which the parties are agreed; and why the parties are correct in their agreement.
History
Sections 8 and 9 deal with different forms of criminal responsibility as an accessory before the fact. The origin of the expression "a probable consequence" in those sections can be traced to Sir Matthew Hale, who said[14]:
"[I]f A. command B. to beat C. and B. beats C. so that he dies, A. is accessary, because it may be a probable consequence of his beating ... [T]he like it is if he command B. to rob him, and in robbing him B. kills him, A. is accessary to the murder."
[14]Hale, Historia Placitorum Coronae, (1736), vol 1 at 617 (emphasis added).
The third edition (1792) of Foster's Crown Law, after referring to Hale, dealt as follows with "cases where a person supposed to commit a felony at the instigation of another hath gone beyond the terms of such instigation, or hath, in the execution, varied from them"[15]. It said[16]:
[15]Foster, A Report of Some Proceedings on the Commission for the Trial of the Rebels in the Year 1746, in the County of Surry; and of Other Crown Cases, 3rd ed (1792) at 369.
[16]Foster, A Report of Some Proceedings on the Commission for the Trial of the Rebels in the Year 1746, in the County of Surry; and of Other Crown Cases, 3rd ed (1792) at 369 (emphasis in original).
"If the principal totally and substantially varieth, if being solicited to commit a felony of one kind he wilfully and knowingly committeth a felony of another, he will stand single in that offence, and the person soliciting will not be involved in his guilt. For on his part it was no more than a fruitless ineffectual temptation. The fact cannot with any propriety be said to have been committed under the influence of that temptation."
It continued[17]:
[17]Foster, A Report of Some Proceedings on the Commission for the Trial of the Rebels in the Year 1746, in the County of Surry; and of Other Crown Cases, 3rd ed (1792) at 369.
"But if the principal in substance complieth with the temptation, varying only in circumstance of time or place, or in the manner of execution, in these cases the person soliciting to the offence will, if absent, be an accessary before the fact, if present a principal."
Then it said[18]:
[18]Foster, A Report of Some Proceedings on the Commission for the Trial of the Rebels in the Year 1746, in the County of Surry; and of Other Crown Cases, 3rd ed (1792) at 370 (emphasis in original).
"So where the principal goeth beyond the terms of the solicitation, if in the event the felony committed was a probable consequence of what was ordered or advised, the person giving such orders or advice will be an accessary to that felony."
Dixon and Evatt JJ quoted this passage in Brennan v The King[19]. The expression "a probable consequence" is to be noted. The work then gave three examples[20]:
[19](1936) 55 CLR 253 at 263. Dixon and Evatt JJ cited the passage as being from the 1809 edition. Foster's Crown Law is often referred to in the 1809 edition. In fact there was no 1809 edition. There was, rather, in that year a reprint of the 3rd edition of 1792.
[20]Foster, A Report of Some Proceedings on the Commission for the Trial of the Rebels in the Year 1746, in the County of Surry; and of Other Crown Cases, 3rd ed (1792) at 370.
"A., upon some affront given by B., ordereth his servant to way-lay him and give him a sound beating; the servant doth so, and B. dieth of this beating. A. is accessary to this murder.
A. adviseth B. to rob C., he doth rob him, and in so doing, either upon resistance made, or to conceal the fact, or upon any other motive operating at the time of the robbery, killeth him. A. is accessary to this murder.
Or A. soliciteth B. to burn the house of C.; he doth it; and the flames taking hold of the house of D. that likewise is burnt. A. is accessary to the burning of this latter house."
The work concluded[21]:
"These cases are all governed by one and the same principle. The advice, solicitation, or orders in substance were pursued, and were extremely flagitious on the part of A. The events, though possibly falling out beyond his original intention, were in the ordinary course of things the probable consequences of what B. did under the influence, and at the instigation of A. And therefore, in the justice of the law, he is answerable for them."
Stephen J said in Johns v The Queen that the examples given reveal that the author cannot have meant by "a probable consequence" a consequence that was more probable than not[22].
[21]Foster, A Report of Some Proceedings on the Commission for the Trial of the Rebels in the Year 1746, in the County of Surry; and of Other Crown Cases, 3rd ed (1792) at 370 (emphasis in original).
[22](1980) 143 CLR 108 at 120.
In 1877, Sir James Fitzjames Stephen published A Digest of the Criminal Law (Crimes and Punishments). It contained no clear precursor to s 8 of the Code, but Art 41 was a precursor to s 9[23]. The title was "Where crime committed is probable consequence of crime suggested." Despite the word "probable" in that title, Art 41 provided[24]:
"If a person instigates another to commit a crime, and the person so instigated commits a crime different from the one which he was instigated to commit, but likely to be caused by such instigation, the instigator is an accessory before the fact."
[23]cf Johns v The Queen (1980) 143 CLR 108 at 128, where Mason, Murphy and Wilson JJ suggested that Art 41 was a precursor to s 8, and Art 38 to s 7. Article 38 bore the title "Common purpose" and provided:
"When several persons taken part in the execution of a common criminal purpose, each is a principal in the second degree, in respect of every crime committed by any one of them in the execution of that purpose.
If any of the offenders commits a crime foreign to the common criminal purpose, the others are neither principals in the second degree, nor accessories unless they actually instigate or assist in its commission."
[24]Stephen, A Digest of the Criminal Law (Crimes and Punishments), (1877) at 25-26.
In Johns v The Queen, Stephen J suggested that "likely" did not here appear to mean "more probable to happen than not", since among the illustrations given by Sir James Fitzjames Stephen was the second example given in Foster's Crown Law[25].
[25](1980) 143 CLR 108 at 120.
Sir James Fitzjames Stephen then suggested to the Lord Chancellor (Lord Cairns) and the Attorney-General (Sir John Holker) that the Digest could be made into a draft Penal Code. They authorised him to prepare a draft Penal Code, and this he did. Sir John Holker introduced it into Parliament as the Criminal Code Indictable Offences Bill 1878. According to Stephen[26]:
"[T]he bill was favourably received, but Parliament had not time to attend to it. A commission, however, was issued to Lord Blackburn, Mr Justice Barry, Lord-Justice Lush, and myself, to inquire into and consider and report upon the Draft Code. It was accordingly considered by us for about five months, namely from November, 1878, to May, 1879. We sat daily during nearly the whole of that time, and discussed every line and nearly every word of every section. The Draft Code which was appended to the Report speaks for itself. It differs slightly from the Draft Code of 1878. The particulars of the differences are stated in the Report prefixed to the Draft Code of 1879."
[26]Stephen, A History of the Criminal Law of England, (1883), vol I at vi.
The relevant provisions of Stephen's draft Penal Code were ss 71 and 72, which were part of Ch IV. They corresponded with ss 71 and 72 in Pt IV of the Draft Code of 1879 which was appended to the Report. The Report said of Ch IV and Pt IV: "Each effects a change, not so much in the substance as in the language of the existing law."[27]
[27]United Kingdom, Report of the Royal Commission Appointed to Consider the Law Relating to Indictable Offences, (1879) [C 2345] at 19.
The concluding words of s 71 of the Draft Code of 1879 were:
"If several persons form a common intention to prosecute any unlawful purpose, and to assist each other therein, each of them is a party to every offence committed by any one of them in the prosecution of such common purpose, the commission of which offence was or ought to have been known to be a probable consequence of the prosecution of such common purpose."
The concluding words of s 72 were:
"Every one who counsels or procures another to be a party to an offence is a party to every offence which that other commits in consequence of such counselling or procuring, and which the person counselling or procuring knew or ought to have known to be likely to be committed in consequence of such counselling or procuring."
The marginal note to s 72 stated: "This is believed to express the existing law. See Foster, Chapter III." As noted above, Foster's Crown Law did not describe the offence committed as being one which "the person counselling or procuring knew or ought to have known to be likely to be committed in consequence of such counselling or procuring". Instead it described the relevant events as being "in the ordinary course of things the probable consequences of what B. did under the influence, and at the instigation of A.".
The Draft Code of 1879 was put to Parliament as the Criminal Code Bill in 1880. Section 71 of the Draft Code of 1879 became s 72 of the Criminal Code Bill 1880 and s 72 of the Draft Code of 1879 became s 73 of the Criminal Code Bill 1880. The Criminal Code Bill 1880 was not enacted, in part because the government changed in 1880, and in part because of Cockburn CJ's criticism of it[28]. However, the Bill aroused interest in Canada, New Zealand and Australia. This led to legislation in Canada in 1892. It led to legislation in New Zealand in 1893. It led to the preparation in 1897 by Sir Samuel Griffith for the Government of Queensland of a Draft Code of Criminal Law. That Draft Code was enacted in Queensland in 1899 and in Western Australia in 1902. A version of Sir Samuel's Draft Code was enacted in Tasmania in 1924 and in the Northern Territory in 1983.
[28]Radzinowicz, Sir James Fitzjames Stephen 1829-1894 and his Contribution to the Development of Criminal Law, (1957) at 20-21.
Section 10 of Sir Samuel Griffith's Draft Code was identical to s 8 of the present Code (and to s 8(1) of the Criminal Code (WA) and s 4 of the Criminal Code (Tas)), and s 11 was identical to s 9(1) of the present Code (and to s 9 of the Criminal Code (WA) and s 5 of the Criminal Code (Tas) (substantially))[29].
[29]The Criminal Code (NT) took a different course: s 8(1) and s 9 each render the alleged accessory liable unless the alleged accessory proves that he or she did not foresee the commission of the offence as "a possible consequence" of prosecuting the unlawful purpose or giving the counsel respectively.
Sir Samuel attached the following marginal note to s 10 of his Draft Code: "Common Law. Bill of 1880, s 72." This type of note would normally indicate that he viewed the provision as preserving the common law. Section 10 of Sir Samuel's Draft Code did not correspond with the common law as it is now understood to be, namely that the principal offender's crime must be foreseen by the accessory "as a possible incident of the common unlawful enterprise"[30]. It did, however, bear some resemblance to the statement of the common law to be found in the 1896 edition of Russell[31]:
"It is submitted that the true rule of law is, that where several persons engage in the pursuit of a common unlawful object, and one of them does an act which the others ought to have known was not improbable to happen in the course of pursuing such common unlawful object, all are guilty."
Nor did s 10 of Sir Samuel's Draft Code correspond with s 72 of the Criminal Code Bill of 1880, since it omitted the words "or ought to have been known to be". In this respect it differed from s 61 of the Criminal Code 1892 (Can), which contained a provision identical to s 72; the corresponding but different provision now in force is s 21(2) of the Criminal Code (Can) (using the phrase "... knew or ought to have known that the commission of the offence would be a probable consequence of carrying out the common purpose"). In this respect s 10 of Sir Samuel's Draft Code also differs from the New Zealand legislation enacted successively as the Criminal Code Act 1893, s 73 and the Crimes Act 1908, s 90(2) (using the phrase "... was or ought to have been known to be a probable consequence"); the legislation is now in force in a different form as s 66(2) of the Crimes Act 1961 (using the phrase "... known to be a probable consequence of the prosecution of the common purpose").
[30]Chan Wing-Siu v The Queen [1985] AC 168 at 175.
[31]Russell, A Treatise on Crimes and Misdemeanors, 6th ed (1896), vol I at 169 (emphasis added). See also McAuliffe v The Queen (1995) 183 CLR 108 at 114-116 per Brennan CJ, Deane, Dawson, Toohey and Gummow JJ.
Sir Samuel Griffith also placed the following marginal note to s 11 of his Draft Code: "Common Law. Bill of 1880, s 73." It is true that s 11 did correspond substantially with the common law as reflected in Foster's Crown Law and, for example, in Halsbury's Laws of England, published in 1909[32]. It is not, however, true that s 11 corresponded exactly with s 73 of the Criminal Code Bill of 1880: it adopted a "probable consequence" test in lieu of the words "knew or ought to have known to be likely to be committed".
[32]Halsbury's Laws of England, 1st ed, vol IX, "Criminal Law and Procedure", par 532, at 255: "Where the principal goes beyond the terms of the instigation, yet if the advice or order of the instigator is substantially followed or obeyed, and the felony committed was a probable consequence of what was ordered or advised, the person giving the order or advice is an accessory to that felony."
The history thus briefly set out does not offer any decisive reason for selecting one rather than another construction for "a probable consequence" in ss 8 and 9 of the Code. But it does serve as a reminder of how great the difference is between s 8 of the Code and the common law as now understood, for the common law stress on the need for "a crime foreseen as a possible incident of the common unlawful enterprise"[33] is quite different from Sir Samuel Griffith's requirement of a probable consequence. The history illustrates how Sir Samuel Griffith decided to adopt the entirely objective approach in Foster's Crown Law for the problem addressed by s 9 and extend it to the problem addressed by s 8. And the history also suggests that the failure of Sir Samuel's Draft Code to take up as an element in any test the mental state of the accused, or what the accused ought to have known, meant that very wide liability would potentially rest on accused persons if "a probable consequence" were given a meaning extending to any possibility which might be described as real or substantial.
[33]Chan Wing-Siu v The Queen [1985] AC 168 at 175 (emphasis added).
The language of the Code generally
The word "probable" was only used in the original Code in s 8 and s 9(1). It is now also used in s 10A(2), a provision introduced in 1997. Apart from those three instances, it is used in only one place in the Code. That is s 415(1)(a), dealing with the crime of extortion, where the expression "without reasonable or probable cause" is employed. Like s 10A(2), s 415 did not appear in the original Code. The word "likely" is used in many parts of the Code – some appearing in the original Code, some added since. But neither s 415 nor the sections using the word "likely" cast any light on the meaning of "a probable consequence" in s 8 or s 9.
Authorities relied on by Fitzgerald P and Pincus JA
The only prior authority of this Court on the meaning of "a probable consequence" in s 8 or its equivalents is Brennan v The King[34], a decision on s 8 of the Criminal Code (WA), which was in the same terms as s 8 of the Code[35]. It is convenient to set out two passages – one from Starke J's reasons for judgment and one from those of Dixon and Evatt JJ – together with the comments on those passages made by Fitzgerald P in R v Hind and Harwood[36].
[34](1936) 55 CLR 253.
[35]Stuart v The Queen (1974) 134 CLR 426 discussed ss 8 and 9, but not this aspect of them.
[36](1995) 80 A Crim R 105 at 116-117.
Starke J said[37]:
"A probable consequence is, I apprehend, that which a person of average competence and knowledge might be expected to foresee as likely to follow upon the particular act; though it may be that the particular consequence is not intended or foreseen by the actor. This is not a definition, but 'only a guide to the exercise of common sense'. Now if a person commits manslaughter who brings about the death of another by some unlawful act, then it must be taken, I think, that death is treated in law as a not improbable consequence of such an act, either because of the definition of the crime or because experience has established that such a result ought to be foreseen and expected. Under a proper charge, therefore, a verdict of manslaughter against the prisoner Brennan could upon the evidence be supported."
Fitzgerald P commented[38]:
"The overall effect of the passages emphasised is that it is sufficient to satisfy s 8 if a consequence was a real or substantial possibility; indeed, the second passage appears to assert that probability can be proved by linking cause and effect."
[37](1936) 55 CLR 253 at 260-261. The emphasis is that given by Fitzgerald P.
[38]R v Hind and Harwood (1995) 80 A Crim R 105 at 116.
Dixon and Evatt JJ said[39]:
"[D]eath can be considered the probable consequence of the prosecution of the purpose if the purpose in which [Brennan] concurred made it likely that his confederates would, if necessary, use violence and such a kind or degree of violence as would probably cause death. The fact that, according to the verdict, they must be taken to have used an amount of force less than might have been contemplated by them, would not make the death, which ordinarily would not, but actually did, result from such a use of force, too remote a consequence for the prosecution of the purpose."
[39](1936) 55 CLR 253 at 264. The emphasis is that given by Fitzgerald P.
Fitzgerald P commented[40]:
"Again, the tenor of these statements is that it is sufficient to satisfy s 8 if there was a real possibility that an offence of the nature of the offence committed would be committed."
[40]R v Hind and Harwood (1995) 80 A Crim R 105 at 117.
The statements by Starke J and by Dixon and Evatt JJ explain "a probable consequence" in terms of likelihood, without explaining which degree of likelihood is relevant. For that reason alone it is not possible to agree with Fitzgerald P that the overall effect of the statements is that s 8 is satisfied if a particular consequence is a real possibility or a substantial possibility. As the appellants submitted, there is no detectable basis in Brennan v The King for that point of view.

Fitzgerald P[41] also discussed Johns v The Queen[42], but that was a case on accessories before the fact at common law, and in any event, so far as it discussed "probable", it did not define it, except to contrast it with "possible".
[41]R v Hind and Harwood (1995) 80 A Crim R 105 at 117-119.
[42](1980) 143 CLR 108.
Pincus JA was of opinion in R v Hind and Harwood that "likely" in s 302(1)(b) of the Code conveyed "the idea that the act in question created a substantial or real chance of danger to human life, regardless of whether that chance was more or less than 50 per cent"[43]. The meaning he gave to "likely" in s 302(1)(b) was based on four matters.
[43]R v Hind and Harwood (1995) 80 A Crim R 105 at 141.
One was the construction given by this Court in Boughey v The Queen to the words "likely to cause death" in the provision in the Criminal Code (Tas) which is equivalent to s 302(1)(b) of the Code, namely, s 157(1)(c)[44]. He pointed out that the majority in that case said that "likely to cause death" conveyed "the notion of a substantial – a 'real and not remote' – chance regardless of whether it is less or more than 50 per cent"[45]. The second matter was the meaning of the word "probable" in the test for malice aforethought at common law[46]:
"If an accused knows when he does an act that death or grievous bodily harm is a probable consequence, he does the act expecting that death or grievous bodily harm will be the likely result, for the word 'probable' means likely to happen."
The third matter referred to by Pincus JA[47] was that in Brennan v The King:
"Dixon and Evatt JJ in discussing s 8 adopted the test whether the purpose concurred in 'made it likely that [Brennan's] confederates would, if necessary, use violence and such a kind or degree of violence as would probably cause death'[48]. The notion that one is entitled to add a contingency ('if necessary') which may not itself be probable is implicit in this passage."
And the fourth matter referred to by Pincus JA[49] was the judgment of Gibbs J in Stuart v The Queen[50]. Pincus JA said that Gibbs J "rejected the notion that the question is whether 'viewed a priori, murder is a possible consequence of extortion'". Pincus JA then quoted Gibbs J's later statement: "It is not uncommon for deaths to occur when a fire breaks out in a building containing people." Pincus JA said of this language that it is "consistent with an equation of 'probable' with 'likely' and with the notion that each implies that there is a substantial or real chance."[51]
[44](1995) 80 A Crim R 105 at 137-138, 141.
[45]Boughey v The Queen (1986) 161 CLR 10 at 21. Pincus JA also relied on the equation by Gibbs CJ of "likely" and "probable" – perplexingly, since Gibbs CJ disagreed with the majority on the question of whether a jury direction could be given using the word "chance".
[46](1995) 80 A Crim R 105 at 141, quoting R vCrabbe (1985) 156 CLR 464 at 469 per Gibbs CJ, Wilson, Brennan, Deane and Dawson JJ. Pincus JA also referred to the fact that in Boughey v The Queen (1986) 161 CLR 10 at 19 Mason, Wilson and Deane JJ pointed out that in R v Crabbe at 469-470 "probable" and "likely" were used as synonyms.
[47](1995) 80 A Crim R 105 at 141-142.
[48](1936) 55 CLR 253 at 264 (Pincus JA's emphasis).
[49](1995) 80 A Crim R 105 at 142.
[50](1974) 134 CLR 426 at 443 (Pincus JA's emphasis).
[51](1995) 80 A Crim R 105 at 142.
These considerations do not support the conclusion that "a probable consequence" in s 8 means no more than "a substantial or real chance".
As to the first point, authorities on the construction of the words "likely to cause death" in s 157(1)(c) of the Criminal Code (Tas) are not decisive on the meaning of the quite different words "a probable consequence" in s 8 or s 9 of the Code. The former words relate to the potential responsibility for murder as a principal offender of someone who has unlawfully killed another in the prosecution of an unlawful purpose. The latter words relate to the potential responsibility otherwise than as a principal offender for conduct which was not within a common purpose or a counselled offence, but which is only a consequence of the common purpose or the counselled offence. These differences in context and language suggest a construction which would make the test created by the former words – "likely to cause death" – easier to satisfy than the test created by the latter words – "a probable consequence"[52].
[52]Furthermore, Gibbs CJ and Brennan J disagreed with the majority view stated in Boughey v The Queen (1986) 161 CLR 10 at 22 per Mason, Wilson and Deane JJ: see at 15 and 42-44.
As to the second point, a common law test for malice aforethought is not necessarily applicable to s 8 or s 9, and the authority cited in any event does not explain what is meant by "likely".
As to the third point, it is not convincing to rely on Dixon and Evatt JJ's use of the words "if necessary" to attribute to them adherence to a meaning for "probable" which they declined explicitly to state.
As to the fourth point, between the two passages from Gibbs J's reasons for judgment in Stuart v The Queen quoted by Pincus JA appeared the following[53]:
"Under s 8 it is necessary for the jury to consider fully and in detail what was the unlawful purpose and what its prosecution was intended to entail and what was the nature of the actual crime committed, and then to decide whether that crime was of such a nature that its commission was a probable consequence of the prosecution of that purpose."
Gibbs J then analysed the facts of the particular case in detail. It was as part of that analysis that the second passage quoted by Pincus JA appears. In those circumstances it does not follow that Gibbs J was asserting that "not uncommon" should be treated as being equivalent to "likely" and "probable".
[53](1974) 134 CLR 426 at 443.
The authorities relied on in the Queensland Court of Appeal do not support the conclusion at which it arrived and which trial courts in that State have been applying. Is it possible nonetheless to justify the conclusion as a matter of principle?
"A probable consequence" is not a consequence more likely to have happened than not
The parties agreed that "a probable consequence" did not mean a consequence likely to happen on the balance of probabilities. They were correct for several reasons.
One is that in ss 8 and 9(1) the expression is "a probable consequence", not "the probable consequence". Had the expression been "the probable consequence", it would have pointed towards a balance of probabilities test. The expression actually used points against that test. The expression "a probable consequence" is compatible with there being more inconsistent probable consequences than one resulting from the prosecution of a particular purpose or the carrying out of particular counsel; where there are a number of inconsistent "probable" consequences it is difficult to see how all can be more probable than not.
A second reason is that if a balance of probabilities test applied, it would mean that accused persons could escape conviction if the prosecution failed to do more than demonstrate that the risk of the consequence described in ss 8 and 9 was plainly there and that the odds were only just against it[54]. Yet that outcome would appear to be contrary to the structure of the two sections, the function of which is to widen criminal responsibility. That is, the objects of ss 8 and 9 could be frustrated if "a probable consequence" of an event meant "a consequence which, judged from the time of the event, was more probably likely to happen than not"[55].
[54]See R v Piri [1987] 1 NZLR 66 at 78 per Cooke P, McMullin and Somers JJ concurring.
[55]See, for example, R v Gush [1980] 2 NZLR 92 at 95 per Richmond P, Richardson and O'Regan JJ; R v Hagen, Gemmell and Lloyd unreported, Court of Appeal of New Zealand, 4 December 2002 at [46] per Tipping, McGrath and Anderson JJ.
Finally, as counsel for the second appellant submitted: "The Code could easily read 'a consequence that was more probable than not'. It does not."
"A probable consequence" is not a consequence which is barely possible
The respondent did not contend for the view that "a probable consequence" referred to one which was no more than barely possible. In that it was correct. Under the common law rule dealing with the same problem as s 8, criminal responsibility depends "upon the jury's assessment of whether or not the accessory before the fact must have been aware of the possibility that responses by the victim or by third parties would produce the reaction by the principal offender which led to the [crime which the principal offender committed]"[56]. Thus at common law the test is "subjective", not "objective". It depends upon the defendant's awareness of possibilities, and not on the existence, independently of the defendant's awareness, of probabilities. The selection by the legislature of an objective, not a subjective test, and the selection of the language of probability, not possibility (let alone bare possibility), point towards a construction of ss 8 and 9 as excluding the sufficiency of bare possibilities.
[56]Johns v The Queen (1980) 143 CLR 108 at 118 per Stephen J (emphasis in original); see also at 130-131 per Mason, Murphy and Wilson JJ.
"A probable consequence" means more than a "real possibility or chance"
The respondent accepted that "a probable consequence" meant more than a consequence which was reasonably possible or which had a reasonable chance of coming to pass. That stance was sound, because ss 8 and 9 mark a contrast between establishing what is probable and what is possible, and the contrast is significant. The reasons why that is so are discussed below[57].
[57]See at [72]-[81].
It is now necessary to examine the arguments of the parties on the points about which they disagreed.
Did the trial judge err in explaining the meaning of "a probable consequence"?
The first and second appellants submitted that the trial judge had erred in saying anything about the meaning of "a probable consequence". They contended that the expression required no elaboration, being well understood by ordinary people[58], and that attempts to elaborate it are likely to lead to confusion. They referred to a statement by Sir Samuel Griffith: "A Code ought, if possible, to be so framed as to require no definitions of terms in common use in ordinary speech or writing."[59]
[58]Relying on R v Salmon & James [2003] QCA 17 at [45] per McMurdo P, Jerrard JA concurring at [69] and Helman J at [70].
[59]"An Explanatory Letter to the Honourable the Attorney-General", in Draft of a Code of Criminal Law prepared for the Government of Queensland, (1897) at viii.
They also relied on various statements in the authorities. One was a statement by Kirby J in a case concerning s 23 of the Code. He said that the provisions of the Code "should be capable of being explained to a jury, according to their own terms, which (at least in the present connection) are relatively simple in their expression"[60]. Another statement on which the first and second appellants relied was that of Mason, Wilson and Deane JJ in Boughey v The Queen[61]:
"A basic objective of any general codification of the criminal law should be, where practicable, the expression of the elements of an offence in terms which can be comprehended by the citizen who is obliged to observe the law and (where appropriate) by a jury of citizens empanelled to participate in its enforcement ... The courts should, however, be wary of the danger of frustrating that basic purpose of codification of the criminal law by unnecessarily submerging the ordinary meaning of a commonly used word in a circumfluence of synonym, gloss and explanation which is more likely to cause than to resolve ambiguity and difficulty."
The first and second appellants relied on a model direction given by the Western Australian Court of Criminal Appeal in relation to liability under the Western Australian equivalents of s 8 and s 302(1)(b)[62], which contained no elaboration of the meaning of "probable".
[60]Murray v The Queen (2002) 211 CLR 193 at 218 [78].
[61](1986) 161 CLR 10 at 21.
[62]R v Seiffert and Stupar (1999) 104 A Crim R 238 at 247-248 per Pidgeon J, Kennedy and White JJ concurring.
Finally, the first and second appellants submitted that "a probable consequence" was an expression like "beyond reasonable doubt", which this Court has repeatedly said is not to be elaborated upon[63].
[63]For example, Thomas v The Queen (1960) 102 CLR 584 at 595 per Kitto J; Dawson v The Queen (1961) 106 CLR 1 at 18 per Dixon CJ; Green v The Queen (1971) 126 CLR 28 at 32-33 per Barwick CJ, McTiernan and Owen JJ.
The submission that the trial judge erred in saying anything about the meaning of "a probable consequence" is to be rejected. In R v Piri, Cooke P said, correctly with respect, in rejecting a similar argument[64]:
"With some ordinary English words that is a feasible and accepted approach. And with the words 'likely' and 'probable' there are occasions when it is unnecessary for the Judge to expand on their meaning; the Judge can simply leave the case to the jury without elaboration in this respect. But where a critical issue as to the degree of likelihood or probability clearly arises, that may not do. The jury may then be entitled to more guidance".
There is also Canadian authority that it is permissible to explain the meaning of a provision similar (but not identical) to s 8, namely, s 21(2) of the Criminal Code (Can)[65].
[64][1987] 1 NZLR 66 at 79, McMullin and Somers JJ concurring.
[65]R v Cribbin (1994) 89 CCC (3d) 67 at 77 per Morden ACJO, Catzman and Arbour JJA.
The expression "a probable consequence" consists of ordinary English words, but they have no single meaning common to lay speakers. Acceptance of the argument advanced by the first and second appellants would mean that trial judges would be precluded from answering questions posed by jurors about the meaning of the expression "a probable consequence". In a given case a jury might understandably experience difficulties with the expression "a probable consequence" while in another case a different jury may not. A rule of law which banned judicial attempts to deal with these difficulties would be perverse. Acceptance of the argument advanced by the first and second appellants would also create a formulaic approach to summing up. The function of a summing up is to furnish information which will help a particular jury to carry out its task in the concrete circumstances of the individual case before it and in the light of the trial judge's assessment of how well that jury is handling its task. It is undesirable for a summing up to assume the character of a collection of hallowed phrases mechanically assembled on a priori principles to be mouthed automatically in all circumstances, whether or not a particular jury actually understands them. If a judge sees it as desirable to explain the meaning of "a probable consequence" – perhaps because it is perceived that some jurors may think it calls for a balance of probabilities analysis while others may think it refers to relatively remote possibilities – the Code does not prevent this from being done.
To reject the argument advanced by the first and second appellants is not to depart from the views of Sir Samuel Griffith or the statements in this Court quoted earlier. Sir Samuel was speaking of how a Code should be framed, not of how a jury should be directed. Kirby J's observations in Murray v The Queen were directed to "relatively simple" provisions: the meaning of "a probable consequence" is not relatively simple, but differs with context. The observation of Mason, Wilson and Deane JJ in Boughey v The Queen about drafting Codes so as to make them comprehensible to juries was qualified by the words "where practicable", and their warning about not submerging ordinary meanings of commonly used words was a warning only against doing so "unnecessarily": it is not always practicable to avoid explanations of "a probable consequence" and it may be necessary to give them. Indeed, in that very case, Mason, Wilson and Deane JJ, after quoting part of the trial judge's summing up there under challenge in which he offered an account of what "likely to cause death" meant, said that the passage "contained helpful and correct guidance for the jury" in making the point that it "is an ordinary expression which is meant to convey the notion of a substantial or real chance as distinct from what is a mere possibility: 'a good chance that it will happen'; 'something that may well happen'; something that is 'likely to happen'"[66]. Further, the fact that in some cases, such as R v Seiffert and Stupar[67], a direction contained no explanation of "probable" does not establish that other directions which do seek to explain it are on that ground alone erroneous.
[66]Boughey v The Queen (1986) 161 CLR 10 at 22.
[67](1999) 104 A Crim R 238 at 247-248 per Pidgeon J, Kennedy and White JJ concurring.
The stand which this Court has taken on the expression "beyond reasonable doubt" – that it alone must be used, and nothing else – has not been shared elsewhere[68]. Even in Australia it is an extreme and exceptional stand. The justification for it rests on several considerations. One is that "beyond reasonable doubt" is an expression "used by ordinary people and is understood well enough by the average man in the community"[69]. That is not so of "a probable consequence". A second consideration is that departures from the formula "have never prospered"[70]. That has not been demonstrated to be the case in relation to "a probable consequence". A third consideration is that expressions other than "beyond reasonable doubt" invite the jury "to analyse their own mental processes"[71], which is not the task of a jury. "They are both unaccustomed and not required to submit their processes of mind to objective analysis"[72]. Explanation of the expression "a probable consequence" does not require this of juries. Finally, as Kitto J said in Thomas v The Queen[73]:
"Whether a doubt is reasonable is for the jury to say; and the danger that invests an attempt to explain what 'reasonable' means is that the attempt not only may prove unhelpful but may obscure the vital point that the accused must be given the benefit of any doubt which the jury considers reasonable."
There is not in that respect any analogy between "beyond reasonable doubt" and "a probable consequence".
[68]Walters v The Queen [1969] 2 AC 26 at 29 per Lord Diplock; Ferguson v The Queen [1979] 1 WLR 94 at 99; [1979] 1 All ER 877 at 882 per Lord Scarman (in general it is wise to employ "beyond reasonable doubt", but "other words will suffice, so long as the message is clear"). In New Zealand it is permissible to elaborate on the words to some extent (R v Harbour [1995] 1 NZLR 440 at 448 per Richardson, Casey, Hardie Boys, McKay and Tompkins JJ), as by contrasting a reasonable doubt with a "vague or fanciful doubt": R v Speakman (1989) 5 CRNZ 250 at 260. To similar effect are cases in Canada (R v Lifchus [1997] 3 SCR 320 at 327-330 [15]-[22] per Lamer CJ, Sopinka, Cory, McLachlin, Iacobucci and Major JJ (La Forest, L'Heureux-Dubé and Gonthier JJ concurring)) and the United States (Victor v Nebraska 511 US 1 at 26 (1994) per Ginsburg J).
[69]Dawson v The Queen (1961) 106 CLR 1 at 18 per Dixon CJ; Green v The Queen (1971) 126 CLR 28 at 31 per Barwick CJ, McTiernan and Owen JJ.
[70]Dawson v The Queen (1961) 106 CLR 1 at 18 per Dixon CJ.
[71]Thomas v The Queen (1960) 102 CLR 584 at 606 per Windeyer J.
[72]Green v The Queen (1971) 126 CLR 28 at 33 per Barwick CJ, McTiernan and Owen JJ.
[73](1960) 102 CLR 584 at 595.
Did the trial judge err in saying more than that "probable" meant "likely"?

The second appellant in particular then fell back on an argument that, if the trial judge was entitled to say anything, he was entitled only to say that "probable consequence" meant "likely consequence". To go further would involve using other phrases to explain "probable" and "likely" which themselves would tend to multiply the risks of confusion and to undercut the benefits at which codification was aimed.
The flaw in this argument is that just as the range of possible meanings for "probable" may justify some explanation of the expression, to offer as an explanation that the word means "likely" is only to point to a further word which also carries diverse meanings.
What directions may and may not be given?
The first and third appellants did not explicitly advance any argument beyond the two just examined. But implicit in their position was a further contention, which eventually was explicitly advanced on behalf of the second appellant: that even if the trial judge did not err in commenting at all on the meaning of "a probable consequence", and even assuming he did not err in going beyond an equation of "probable" and "likely", what he actually said was erroneous. It was erroneous in that he failed to steer a course between saying that a probable consequence was one which was more likely to occur than not (which would have been unduly generous to the appellants), and saying that a probable consequence was a real or substantial possibility or chance (which he in fact said, and which was unduly harsh to the appellants).
This criticism is sound, for the following reasons.
First, the context in which the expression "a probable consequence" appears must be borne in mind. While it is true that ss 7, 8 and 9 apply to many offences other than murder, the crime charged here was murder. Section 305(1) of the Code provides that conviction carries a mandatory and single penalty, the highest known to the law – life imprisonment. Further, the form of liability for murder under consideration is accessorial. Accessorial liability is old, but it is an exception to the general rules of criminal responsibility. Persons liable under s 7, s 8 or s 9 need not be present at the scene of the crimes for which they are convicted, and the fact that those crimes might be committed by the principal offender may never have entered their heads. In construing "a probable consequence" in ss 8 and 9, the extent to which it is likely that Parliament has created strict or vicarious liability in accessories must be considered.
Secondly, the key word in ss 8 and 9 is "probable", not "possible". The word "probable" has diverse meanings, but all common usages of it suggest a more exacting standard than "possible"[74].
[74]See also R v Crabbe (1985) 156 CLR 464 at 469-470 per Gibbs CJ, Wilson, Brennan, Deane and Dawson JJ.
Further, whatever the common law in the late 19th century was in relation to the problem dealt with by s 8 of the Code, it is clear that now at common law an accessory is liable if the principal offender's crime is "foreseen as a possible incident of the common unlawful enterprise"[75]. Although the law has long recognised accessorial liability, it has also long attempted to lay down limits to the accessorial liability of a person who shared a common purpose with a wrongdoer, or who instigated a wrongdoer to commit a crime. The alleged accessory is not to be liable for everything a principal offender did, either vicariously or absolutely. Over time the law has employed different techniques for placing accessorial liability within just limits while continuing to give it substantial room for operation. The common law protects against excessively wide liability by demanding actual foresight, albeit of a possibility. Under ss 8 and 9 of the Code the function of protecting against excessively wide liability turns on the need for probability of outcome, independently of the alleged accessory's state of mind. If under ss 8 and 9 of the Code the expression "a probable consequence" were construed so as to make a possible consequence sufficient, there would be liability in the accessory for whatever the principal offender did, since the fact that the principal offender did it shows that it was possible, and there would be no protection against excessively wide liability.
[75]Chan Wing-Siu v The Queen [1985] AC 168 at 175.
Authority in other jurisdictions with similar legislation stresses the contrast between "probable" and "possible". Thus, the New Zealand Court of Appeal, applying s 66(2) of the Crimes Act 1961, which turns on what is "known to be a probable consequence", could "see no justification for reading 'probable consequence' in s 66(2) as 'possible consequence'"[76]. The same court stressed the need for the summing up to emphasise that the consequences were "probable, not merely possible"[77]. In Canada, too, where s 21(2) of the Criminal Code centres on whether the accessory "knew or ought to have known that the commission of the offence would be a probable consequence" of the carrying out of the unlawful purpose, it is customary to distinguish between what is probable and what is possible[78].
[76]R v Gush [1980] 2 NZLR 92 at 94 per Richmond P, Richardson and O'Regan JJ.
[77]R v Waho unreported, Court of Appeal of New Zealand, 27 April 2005 at [31] per Hammond, Robertson and Potter JJ. See also R v Rapira (2003) 20 CRNZ 396 at 413-414 [53] per Elias CJ, Gault P and McGrath J.
[78]R v Kirkness [1990] 3 SCR 74 at 110 per Wilson and L'Heureux-Dubé JJ (dissenting, but not on this point).
The difficulty in defining "a probable consequence" is that once it is accepted that "probable" does not mean "on the balance of probabilities" and that it means more than a real or substantial possibility or chance, it is difficult to arrive at a verbal formula for what it does mean and for what the jury may be told.
The expression "a probable consequence" means that the occurrence of the consequence need not be more probable than not, but must be probable as distinct from possible. It must be probable in the sense that it could well happen[79].
[79]R v Gush [1980] 2 NZLR 92 at 94 per Richmond P, Richardson and O'Regan JJ; R v Hagen, Gemmell and Lloyd unreported, Court of Appeal of New Zealand, 4 December 2002 at [46] per Tipping, McGrath and Anderson JJ.
In this case, the s 8 question is whether "the offence" – murder by killing the deceased with intent to do some grievous bodily harm – was "a probable consequence" of the prosecution of the common intention of the appellants to prosecute the unlawful purpose of assaulting the deceased. In this case, the s 9 question is whether "the facts constituting the offence actually committed" – the killing of the deceased with intent to do some grievous bodily harm – are "a probable consequence" of carrying out the second appellant's counsel.
It is not necessary in every case to explain the meaning of the expression "a probable consequence" to the jury. But where it is necessary or desirable to do so, a correct jury direction under s 8 would stress that for the offence committed to be "a probable consequence" of the prosecution of the unlawful purpose, the commission of the offence had to be not merely possible, but probable in the sense that it could well have happened in the prosecution of the unlawful purpose. And where it is desirable to give the jury a direction as to the meaning of the expression "a probable consequence" in s 9, a correct jury direction would stress that for the facts constituting the offence actually committed to be "a probable consequence" of carrying out the counselling, they had to be not merely possible, but probable in the sense that they could well have happened as a result of carrying out the counselling.
Did the trial judge's directions comply with the necessary criteria?
The respondent contended that the trial judge's direction in this case did not fail to comply with these criteria. The respondent's argument was that the word "possibility" in the impugned part of the summing up was qualified by the word "real" and the word "chance" by the words "substantial" and "real". On this basis, the respondent submitted that the jury would not have been left with the impression that the appellants could be found guilty in relation to outcomes that were merely possible. That may be true, but the trial judge's direction, compelled by authority as it was, carried the risk of leaving the jury with the impression that the appellants could be found guilty in relation to outcomes which, while more than merely possible, in that they were substantial or real, were not probable. Hence, contrary to the respondent's submissions, the direction that was given by the trial judge was flawed in that it did not convey the idea that the consequence to be looked for was "a probable or likely outcome".
Each appellant has established that the summing up was wrong in law. The point raised by each appeal is thus to be decided in favour of each appellant. The question remains whether the appeals should be dismissed on the ground that "no substantial miscarriage of justice has actually occurred" within the meaning of s 668E(1A) of the Code.
The proviso: general considerations
An appellate court invited to consider whether a substantial miscarriage of justice has actually occurred is to proceed in the same way as an appellate court invited to decide whether a jury verdict should be set aside on the ground that it is unreasonable, or cannot be supported having regard to the evidence. The appellate court must make its own independent assessment of the evidence and determine whether, making due allowance for the natural limitations that exist in the case of an appellate court proceeding wholly or substantially on the record, the accused was proved beyond reasonable doubt to be guilty of the offence on which the jury returned its verdict of guilty[80].
[80]Weiss v The Queen (2005) 80 ALJR 444 at 454-455 [41] per Gleeson CJ, Gummow, Kirby, Hayne, Callinan and Heydon JJ; 223 ALR 662 at 673-674.
A statement of Keane JA in the Court of Appeal in this case provides a starting point. As a matter preparatory to explaining why the second appellant's appeal to the Court of Appeal on the ground that her conviction was unreasonable should be rejected, he said that the nature and extent of the injuries inflicted on the deceased compelled the conclusion that whoever inflicted them did so with the intention of at least doing grievous bodily harm to the deceased[81]. With respect, that is correct. It is now necessary to consider the evidence against each appellant.
[81]R v Deemal-Hall, Darkan & McIvor [2005] QCA 206 at [10].
The proviso: the first appellant
In an initial interview with the police, the first appellant denied any involvement in causing the deceased's death. In a later interview he admitted to the police that while he had given the deceased a "hiding", and was not the only person who had done so, he had used only his bare hands and that it was the third appellant who had used the pickaxe handle. At a further interview he said that the second appellant had told him that she wanted the deceased bashed in the dark and wanted more help, and that the deceased had raped her daughter. He heard the second appellant tell the third appellant to get a weapon, and saw the third appellant go into his house and come out with the pickaxe handle and wearing steel-capped boots. He saw the second appellant hide the pickaxe handle in the van. He observed the deceased's reluctance to leave the van. Before the fight began, the second appellant took the pickaxe handle from the van and hid it behind a bush. She then told the third appellant to get it, and told the first appellant to hit the deceased. The first appellant punched the deceased four times. The deceased fell to the ground, and the first appellant, the third appellant and Bowen began kicking him in the presence of the second appellant. The second appellant called for someone to get the pickaxe handle and, when nobody did, she got it herself. With it she gave the deceased a couple of "good swings" in the back of the neck. The first appellant then departed as the deceased's blood spread "all over the place".
Bowen's evidence suggested that in this account the first appellant had minimised his role, which included kicking the deceased with steel-capped boots and using the pickaxe handle extensively.
Michael James Cobus also implicated the first appellant: his evidence was that the second appellant's attempt to persuade Cobus to "touch up" the deceased was witnessed by the first appellant.
The respondent submitted that, quite independently of s 8 of the Code, the evidence revealed that the first appellant was guilty of murder by reason of either s 7(1)(a) or s 7(1)(c), respectively read with s 302(1)(a). These cases had been left to the jury. The respondent submitted that the first appellant's admissions showed that he intended to do some grievous bodily harm to the deceased, and that he either unlawfully killed the deceased with that intent, or aided the third appellant in unlawfully killing the deceased with that intent. These submissions are correct. An invitation to inflict a bashing at night in a lonely place in company on a man of violent disposition is an invitation intentionally to inflict grievous bodily harm. Whether or not the first appellant used the pickaxe handle, and whether or not it was used as little as he suggested, each punch or kick he inflicted assisted the third appellant in committing murder under s 302(1)(a) because it diminished the will and the ability of the deceased to resist the bashing which the participants had agreed would be given.
Counsel for the first appellant contended that the case against the first appellant depended on Bowen's credibility. Counsel had in mind the fact that Bowen was a witness whose evidence was to be examined with care. Originally he had been charged with murder, but after pleading guilty to a contravention of s 339 of the Code (assault occasioning bodily harm while in company), he received a sentence which was partly suspended and which led to his release after the plea was taken. Bowen accepted that he received a lesser sentence by reason of a promise to give evidence for the prosecution against the three appellants. Counsel submitted that since this Court was not in a position to assess Bowen's credibility, the case was unsuitable for the application of the proviso.
Combining the first appellant's account with the injuries suffered by the deceased, it is clear that the deceased's death was caused by someone who intended to do the deceased grievous bodily harm. The first appellant on his own admission knowingly aided the third appellant, who plainly had the intention of causing grievous bodily harm to the deceased and who must have caused the deceased's death if the first appellant did not. Hence a case of murder beyond reasonable doubt is established against the first appellant. This reasoning does not depend on Bowen's evidence, and it is not weakened by any infirmities in Bowen's credibility.
Counsel for the first appellant also said that questions arose about the reliability of what the first appellant had said to the police. Since the first appellant did not give evidence denying the correctness of what he had said in mechanically recorded interviews, those are not questions, assuming they are capable of arising, which can be answered favourably to the first appellant.
Counsel for the first appellant also contended that since the jury had been misdirected about s 8, the first appellant had not had a trial according to law. The other appellants advanced similar arguments.
In Weiss v The Queen[82] this Court put aside questions relating to two particular kinds of defect in a trial. One was whether the proviso could be applied when there had been "a significant denial of procedural fairness". This does not arise, because the trial was procedurally fair. The other was whether the proviso could be applied where there had been a sufficiently "serious breach of the presuppositions of the trial". This was a reference to a trial which had "so far miscarried as hardly to be a trial at all" or "where an irregularity has occurred which is such a departure from the essential requirements of the law that it goes to the root of the proceedings"[83]. Neither defect existed in relation to the trial so far as it concerned the first appellant.
[82](2005) 80 ALJR 444 at 455 [45]-[46] per Gleeson CJ, Gummow, Kirby, Hayne, Callinan and Heydon JJ; 223 ALR 662 at 675.
[83]Wilde v The Queen (1988) 164 CLR 365 at 373 per Brennan, Dawson and Toohey JJ.
The case against the first appellant was put to the jury on three bases. In relation to one of them, there was a misdirection. The first appellant's admissions made a conviction inevitable on one if not both of the other two bases. There is no reason why the proviso should not be applied in relation to those bases.
Counsel also referred to a statement by Kirby J of the accused's entitlement "to have a jury properly instructed on the elements of the offences charged"[84]. His Honour did not, however, say that in the event of misdirection the proviso could not be applied: his point was rather that it "will often be inappropriate, where misdirection is shown, to invoke a provision such as s 668E(1A) of the Code, even in a strong prosecution case"[85]. His Honour was not there dealing with proceedings which could scarcely be described as trials at all. However, even if it is assumed that some misdirections cannot be overcome by recourse to the proviso, that is not so of this misdirection.
[84]KBT v The Queen (1997) 191 CLR 417 at 433.
[85]KBT v The Queen (1997) 191 CLR 417 at 434.
Proviso: case against the third appellant
The respondent submitted that the third appellant was guilty of murder either because he killed the deceased intending to do him grievous bodily harm, or because his acts aided the first appellant to kill the deceased in the knowledge that the first appellant intended to do him grievous bodily harm.
The third appellant gave the following account to the police. The first appellant told the third appellant that the second appellant wanted them "to smash this fellow up" for money, and that she had told the first appellant: "just come and just bash this lad for me". She asked what a "good spot" would be at which "to get this bloke". The third appellant anticipated that the deceased would suffer a couple of broken bones and be put in hospital. He went home to put on his steel-capped boots with a view to damaging the deceased. The first appellant asked the second appellant what the deceased was like; when she said he was a big man, the first appellant asked the third appellant if he had a stick to use as a weapon. The third appellant obtained the pickaxe handle, and gave it to the first appellant. After the deceased fell to the ground, the third appellant gave him "a little soft kick in the mouth" and kicked him two or three times in the ribs. The first appellant kicked the deceased at the same time. The first appellant asked for the pickaxe handle, and the third appellant supplied it, but evidently fearing that the first appellant would hit the deceased on the head with it, said "Don't kill him" and "Just knock him out". He saw the first appellant hitting the deceased with the pickaxe handle from his feet to his ribs and on the head and "smashing his jaw around" with it; he thought the deceased would suffer a couple of broken bones and a broken jaw.
There was other evidence against the third appellant (apart from Bowen's). His former de facto wife saw him leave home after putting his steel-capped boots on. On his return he showed her a $50 note and the pickaxe handle covered in blood. He told her that he had been in a fight and expected to receive a further $1,000. He told another witness that he had "bashed this guy in the park" with the pickaxe handle.

In the versions of the codes adopted in Canada[105] and in New Zealand[106], the operative provisions require the establishment of even more rigorous elements involving, in each case, a variant of actual knowledge or insight and of probable consequences. Such considerations help to place the exceptional approach of the Griffith Code into the context of the laws applicable elsewhere in Australia and in other countries where statutory codifications and adaptations of the English criminal law are in force. They afford additional reasons for resisting attempts, made by the prosecution in these appeals, to sustain the assimilation of the Code provisions referring to "a probable consequence" to variations on the theme of "possibility" and "chance".
[105]Criminal Code (Can), s 21(2): see joint reasons at [77].
[106]Crimes Act 1961 (NZ), s 66(2): see R v Gush [1980] 2 NZLR 92 at 93-94.
In matters of basic criminal law principle, it is desirable, so far as it can properly be achieved within the statutory language, to ensure uniformity of approach throughout Australia[107]. Complete uniformity on the subject matter of these appeals is impossible, given the content of the common law and the different language of the Australian versions of the Code. However, the insistence on the difference between "probable" and "possible" (or "chance") is certainly a step in the right direction. Where the Code uses the word "probable" it seems unlikely, in this context, that it means "possible". So to construe the Code would not only depart from its text. It would separate even more radically the test applied in Code States from that applicable elsewhere in Australia and indeed in comparable countries overseas.
[107]R v Barlow (1997) 188 CLR 1 at 32; cf Vallance v The Queen (1961) 108 CLR 56 at 75-76.
The erroneous jury direction
The mistaken direction: The joint reasons collect such observations in this Court as throw light on the meaning of the word "probable" appearing in the Code[108]. They explain the adoption by the Court of Appeal of Queensland of the formula used by the trial judge in these proceedings to explain the contested phrase[109]. According to that formulation, it was enough to satisfy ss 8 and 9 of the Code, for example, "if there was a real possibility that an offence of the nature of the offence committed would be committed"[110]. This is how "possibility" and "chance" (qualified by use of the adjective "real") found their way into Queensland trial practice, the Queensland Bench Book and the instructions given to the jury in the present proceedings.
[108]Joint reasons at [42]-[55] referring to Brennan v The King (1936) 55 CLR 253; Stuart v The Queen (1974) 134 CLR 426 at 443; and Boughey v The Queen (1986) 161 CLR 10 at 19, 21-22.
[109]By reference to Hind and Harwood (1995) 80 A Crim R 105 at 116-117.
[110](1995) 80 A Crim R 105 at 117 per Fitzgerald P.
For the reasons given by the other members of this Court, that instruction was erroneous. It did not comply with the language of the Code. Indeed, from internal evidence within the Code, where the word "likely" is quite commonly used[111], it appears clear that the choice of the word "probable" was deliberate[112]. This conclusion is reinforced when the frequency of the general distinction between "probable" and "possible" in legal usage is recognised. It is strengthened still further by contrasting the differential usage of language in the comparable provisions of the codes of different countries, suggesting a deliberate and careful choice of words in the Griffith Code that became the model for Australian codification. It is also reinforced by a reflection on the commonsense conclusion that a decision-maker will find it somewhat more difficult to decide (to the requisite standard) that events are "a probable consequence" of propounded activities than would be the case if all that needs to be shown is that they are a "possibility" or a "chance" (including if those words are qualified by the adjective "real"). Many more consequences are possible, even really possible, than are probable[113]. Reaching a conclusion on possibilities and chances is easier for the mind than having to decide that something is "a probable consequence" of other people's actions and purposes.
[111]The word "likely" is found, inter alia, in the Code, ss 1, 54A(4), 70(1), 71, 75(1)(b), 229J(1), 258(2), 268(1), 269, 270, 271(1), 302(1)(b) and (4), 306(b), 319A(a), 321A(1), 324, 326, 328A(5), 330(1), 331, 332(1), 415(4), 442D(a), 442E, 450E(6), 462(b), 467A(2), 590(2) and (3), 590AP(4), 590AQ(6) and 682(2).
[112]The word "probable" is used only in ss 8, 9, 10A and 415 of the Code.
[113]As an example see Ebner v Official Trustee in Bankruptcy (2000) 205 CLR 337 at 345 [7], 370 [109]; cf Boughey (1986) 161 CLR 10 at 14-15 per Gibbs CJ.
The correct direction: One of the objectives of codification of the criminal law was to avoid unnecessary elaboration of the law. Such elaboration may be prone to confuse rather than to assist juries. Especially where the law has been restated in a code, so as to make a fresh start, it would ordinarily be wrong to gloss the language with notions inherited from the common law or with words that merely represent a judicial attempt, in different language, to restate Parliament's purpose[114].
[114]Murray v The Queen (2002) 211 CLR 193 at 218 [78.1]; Stevens v The Queen (2005) 80 ALJR 91 at 106 [64]; 222 ALR 40 at 57. See also Boughey (1986) 161 CLR 10 at 21; cf joint reasons at [67].
It is not an error for a trial judge, who feels that elaboration to some extent is desirable or necessary, to afford such an elaboration[115]. So much may sometimes be required by jury questions. Indeed, such jury questions often focus upon key expressions, not least the standard formula of "beyond reasonable doubt"[116]. I would not elevate the phrase in question in this appeal to that category so that it may not be elaborated or explained[117]. However, there will often be good sense in simply telling the jury of the requirements of the Code, providing them with a copy of the relevant provisions (as the trial judge did here) and avoiding the use of unnecessary synonyms[118]. Apart from anything else, adhering to the language of the Code tends to narrow the possibility of mistakes and to reduce the length of jury instructions, which are very long in this country, by contrast, say, to jury instructions in the United States of America.
[115]R v Piri [1987] 1 NZLR 66 at 79.
[116]On empirical studies of jurors' ability to follow judicial instructions see Zoneff v The Queen (2000) 200 CLR 234 at 261 [67]; Australian Institute of Judicial Administration, The Jury Project: Stage 1 – A Survey of Australian and New Zealand Judges, (2006) at 30.
[117]See joint reasons at [69] citing Dawson v The Queen (1961) 106 CLR 1 at 18; Green v The Queen (1971) 126 CLR 28 at 31.
[118]See R v Salmon & James [2003] QCA 17 at [45].
Where it is decided that some elaboration of the phrase "a probable consequence" is desirable or necessary, the use of synonyms that invoke the concepts of possibility and chance, however decorated, is seriously wrong. The jury might properly be told that a probable consequence was one that was more than "merely possible"[119]. They might be told that it is one that they would regard as probable in the sense that it "could well have happened" as a result of carrying out the activities in question but they should not be told that it is sufficient if it is proved that what happened was a possibility or real chance or substantial risk.
[119]Cf Boughey (1986) 161 CLR 10 at 15 per Gibbs CJ.
In all of this, I agree with what is said in the joint reasons[120]. The appellants have therefore made out the error for which they contended in these appeals. On the face of things, they are therefore entitled to succeed in their appeals. But their success is subject to the 'proviso'.
[120]See joint reasons at [82]-[83].
The application of the 'proviso'
Terms of the 'proviso': Given that these appeals are decided on the basis of the 'proviso', it is as well to set out the relevant provisions of the Code:
"668E Determination of appeal in ordinary cases
(1)The Court on any such appeal against conviction shall allow the appeal if it is of opinion that … the judgment of the court of trial should be set aside on the ground of the wrong decision on any question of law, or that on any ground whatsoever there was a miscarriage of justice, and in any other case shall dismiss the appeal.
(1A)However, the Court may, notwithstanding that it is of the opinion that the point or points raised by the appeal might be decided in favour of the appellant, dismiss the appeal if it considers that no substantial miscarriage of justice has actually occurred."
The word "Court" in the section means the Court of Appeal[121]. In disposing of these appeals this Court is authorised, on proof of error, to substitute the order which the Court of Appeal ought to have made[122].
[121]The Code, s 668.
[122]Judiciary Act 1903 (Cth), ss 36, 37.
Clearly, from the foregoing, it is established that a "wrong decision on [a] question of law" occurred in the judgment of the court of trial. In the event of a wrong decision, the appellate court is then obliged ("shall") to allow the appeal. The provision for dismissal of the appeal in s 668E(1) is not enlivened. It is only enlivened in terms of s 668E(1A) if, notwithstanding the foregoing opinion, the appellate court comes affirmatively to the conclusion that no substantial miscarriage of justice has actually occurred. So is that the case here?
Lack of appellate consideration: In the conclusion reached by the Court of Appeal, it was unnecessary for that Court to consider, still less apply, the 'proviso'. This was because that Court applied its own authority to the meaning of the word "probable" in the context of ss 8 and 9 of the Code, as expressed in R v Hind and Harwood[123]. It concluded that, in the context, "a probable consequence" was, as the trial judge had instructed the jury, a consequence that is a "substantial risk", "real chance" or "real possibility" of occurring in the prosecution of the common unlawful purpose or in fulfilment of the conduct that was "counselled"[124].
[123](1995) 80 A Crim R 105 at 141.
[124]Cf [2005] QCA 206 at [61].
In this Court, the prosecution urged that comments in the Court of Appeal, addressed to the seriousness of the injuries suffered by the deceased, amounted in some way to an evaluation by that Court of the strength of the respective cases against the appellants. This submission is unsound. In the conclusion it reached, the Court of Appeal had no reason to evaluate the evidence for 'proviso' purposes. Correctly, on its premises, it omitted to do so. The respondent filed no notice of contention in this Court, seeking to support the judgment of the Court of Appeal on the basis of the application of s 668E(1A) of the Code. However, argument proceeded in these appeals on the footing that the 'proviso' was properly before us, by reason of our function. The procedural niceties may be passed by. But the invocation of the 'proviso' in this Court presents many difficulties, for reasons that I will now explain.
Why the 'proviso' is inapplicable
The postulate of a lawful trial: The ordinary postulate of the Australian legal system is that a person, accused of a crime, is entitled to a trial that conforms to the requirements of the law. Most especially, in the trial of serious criminal charges, the person is normally entitled to have the jury, as the "constitutional judge of fact"[125], resolve contested questions of fact by the application of the applicable law correctly explained to them by the presiding judge. Not only is this a basic presumption of the criminal law in this country. It is also a fundamental principle of the international law of human rights. Thus, in Art 14.2 of the International Covenant on Civil and Political Rights ("the ICCPR")[126] it is provided:
"Everyone charged with a criminal offence shall have the right to be presumed innocent until proved guilty according to law." (emphasis added)
[125]Hocking v Bell (1945) 71 CLR 430 at 440.
[126][1980] ATS 23. The Covenant entered into force on 23 March 1976. It entered into force in respect to Australia on 13 November 1980.
A legal mistake in peripheral matters, such as on non-fundamental issues of procedure, an insubstantial error in admitting this or that piece of evidence or a misdirection on a particular point of fact or law arising in the trial may not touch the fundamental requirement of having a trial "according to law"[127]. But where the error that is established involves a mistaken direction with respect to an essential ingredient of the offence and a misdescription to the decision-maker (here the jury) of the content of that ingredient, a real question is presented as to whether the outcome then truly answers to a trial "according to law".
[127]Cf Green v The Queen (1997) 191 CLR 334 at 346-347 per Brennan CJ.
Clearly, the language of the 'proviso' is only enlivened when mistakes have happened. The mistakes which s 668E(1A) contemplates include, explicitly, "the wrong decision on any question of law". However, the 'proviso' is manifestly to be understood against the background of the fundamental assumption that high standards of lawfulness are observed in the conduct of criminal trials. Otherwise, such trials would not only depart from our own high conventional standards but also from the requirements of the ICCPR. Australia has ratified the ICCPR and the First Optional Protocol[128]. This renders the legal standards of this country examinable by the United Nations Human Rights Committee. Inevitably, this consideration brings the influence of the Covenant to bear on the exposition of the applicable Australian law[129]. It suggests that the "wrong decision on any question of law" to which s 668E(1) is addressed is of a particular kind – but one that does not involve fundamental errors on such questions as might deprive the trial of the character of one "according to law"[130].
[128][1991] ATS 39. The First Optional Protocol to the ICCPR entered into force on 23 March 1976. It entered into force in respect of Australia on 25 December 1991.
[129]Mabo v Queensland [No 2] (1992) 175 CLR 1 at 42; Antoun v The Queen (2006) 80 ALJR 497 at 505 [37]-[38]; 224 ALR 51 at 60-61.
[130]See, eg, Wilde (1988) 164 CLR 365 at 373; TKWJ v The Queen (2002) 212 CLR 124 at 148 [76]; Nudd v The Queen (2006) 80 ALJR 614 at 633 [85]; 225 ALR 161 at 184.
Similar remarks, but without reference to the underpinning in human rights law, have been made several times in this Court. The appellate court, deciding the 'proviso' question, is obliged to reach its own conclusion according to the statutory criteria[131]. However, necessarily, it does so in the context of a legal system that observes high standards of compliance with the law; is protective against miscarriages of justice and wrongful convictions; and ordinarily applies the rigorous criterion for proof of criminal guilt, namely proof beyond reasonable doubt[132].
[131]Weiss v The Queen (2005) 80 ALJR 444 at 454 [40]; 223 ALR 662 at 673.
[132]Weiss (2005) 80 ALJR 444 at 454 [39]; 223 ALR 662 at 673.
Defects of trial on the record: There are further considerations to be remembered when the 'proviso' is invoked, as it has been in these proceedings. They include, as this Court pointed out in Weiss v The Queen[133], the limitations inherent in an appellate court conducting a criminal trial on the record. Here that record was taken in the court of trial where that trial, by hypothesis, has miscarried:
"That task is to be undertaken in the same way an appellate court decides whether the verdict of the jury should be set aside on the ground that it is unreasonable, or cannot be supported having regard to the evidence. The appellate court must make its own independent assessment of the evidence and determine whether, making due allowance for the 'natural limitations' that exist in the case of an appellate court proceeding wholly or substantially on the record, the accused was proved beyond reasonable doubt to be guilty of the offence on which the jury returned its verdict of guilty. There will be cases, perhaps many cases, where those natural limitations require the appellate court to conclude that it cannot reach the necessary degree of satisfaction. In such a case the proviso would not apply".
[133](2005) 80 ALJR 444 at 454-455 [41]; 223 ALR 662 at 673-674 (footnotes omitted) (emphasis added). See also Dietrich v The Queen (1992) 177 CLR 292 at 338.
This passage from Weiss provides a salutary reminder both of the appellate duty in criminal appeals and of the necessary limitations within which that duty is performed. In the present proceedings, those limitations have an added feature. This arises because of the fact that, in these appeals, the intermediate appeal court did not perform the function enlivened by the 'proviso'. Effectively, therefore, what is sought by the prosecution in these appeals is a trial of the appellants, on the record, effectively for the first time applying the correct legal standard, and in this Court. In my view, at least in a case of the present kind, the final constitutional and appellate court of the nation is not well equipped to perform such a task.
It is legally authorised to apply the 'proviso'[134]. It may do so although no notice of contention is filed relying on that provision[135] and although it was not considered in the intermediate court[136]. But the existence of the relevant power says nothing about whether that power should be exercised.
[134]Kelly v The Queen (2004) 218 CLR 216 at 238 [56]; Nicholls v The Queen (2005) 219 CLR 196 at 281 [233].
[135]Kelly (2004) 218 CLR 216 at 259 [123]-[126].
[136]Antoun (2006) 80 ALJR 497 at 506-507 [44]-[45], 507 [49], 509 [58]-[60]; 224 ALR 51 at 62-63, 65-66.
Quite apart from the fact that this Court sees no witnesses, is restricted to transcript to which it is taken by the parties with their competing interests, and has inadequate time to consider and absorb all of the evidence viewed in sequence, there are several particular disadvantages in this Court's assuming that role. Doing so represents an aggravated form of the danger, mentioned by McHugh J on several occasions, of "substituting trial by appellate court for trial by jury"[137].
[137]TKWJ (2002) 212 CLR 124 at 148 [76]; cf Nudd (2006) 80 ALJR 614 at 633 [86]; 225 ALR 161 at 184.
In this case, this Court lacks the assistance of the intermediate court's analysis of the evidence. Assuming this function for itself, it deprives the appellants of a right to appeal against the outcome, although effectively it is the appellants' first trial "according to law"[138]. If we misunderstand the evidence; miscalculate the significance of the testimony of Mr Bowen; or misjudge the importance of admissible factual material in a trial that lasted six days and involved some local considerations, we become the source of a miscarriage of justice rather than the source of its correction.
[138]The right to appeal is also a fundamental human right: see ICCPR, Art 14.5; cf Joseph, Schultz and Castan, The International Covenant on Civil and Political Rights: Cases, Materials, and Commentary, 2nd ed (2004) at 452 [14.133].
Final court review of evidence:In the Supreme Court of the United States there have been similar debates over the assumption by that Court of the responsibility to decide the merits of a conviction at trial where legal error in the conduct of a trial is found but it is submitted that such error is "harmless"[139]. In United States v Lane[140], Stevens J, writing in dissent but with the concurrence on this point of Brennan, Marshall and Blackmun JJ, remarked that "[u]ndertaking a harmless-error analysis is perhaps the least useful function that this Court can perform"[141].
[139]Under 28 USC §2111.
[140]474 US 438 (1986).
[141]474 US 438 at 476 (1986); cf United States v Hasting 461 US 499 at 516-518 (1983) (Stevens J, concurring in judgment).

In his reasons Brennan J[142] explained why this was so. He said that, in the nature of the other duties and responsibilities of a final court, it was difficult to perform the "examination of the proceedings in their entirety", the course called for by United States[143] as by equivalent Australian authority[144]. He warned against the temptation to a "perfunctory effort to evaluate the effect of this error"[145] and to the substitution of intuitive judgment and impression for a thorough analysis in "one of the most complex" tasks committed to an appellate court[146]. This is why he concluded the Supreme Court was "manifestly ill-equipped to undertake" such a task so that, if it were to be performed at all, it should normally be remanded to the intermediate court[147]. Much of this analysis has direct application to a request that this Court, for the first time, apply the 'proviso'.
[142]474 US 438 at 464 (1986).
[143]Kotteakos v United States 328 US 750 at 762 (1946).
[144]Driscoll (1977) 137 CLR 517 at 525; Wilde (1988) 164 CLR 365 at 372; Festa (2001) 208 CLR 593 at 632 [122].
[145]474 US 438 at 464 (1986).
[146]474 US 438 at 465 (1986) citing Traynor, The Riddle of Harmless Error, (1970) at 80.
[147]474 US 438 at 465 (1986). In this approach Marshall, Blackmun and Stevens JJ agreed.
The significance of the error: There is a further factor. I now come to why, in these appeals, the erroneous direction given to the jury by the trial judge was so serious. In my estimation, the error of substituting a verbal formula expressed in terms of possibility and chance for the language of the Code, with its reference to "a probable consequence", was factually very significant for the fair trial of the appellants.
There were facts in the evidence that suggest that Ms Deemal-Hall's purpose in counselling the assault upon the deceased did not have murder as a probable consequence of that counselling. Equally, there is evidence which casts doubt on the proposition that a probable consequence of the prosecution of the appellants' common intention to assault the deceased was the deceased's murder. The relevant considerations available to support these propositions included:
(1)Her past close personal association with the deceased manifested in her visit to him the night before his death;
(2)The fact that the deceased voluntarily went with Ms Deemal-Hall to the park where he met his fate. He travelled with her in his van. The precise reason for this conduct is unexplained. However, the inference was available that they met by agreement at Mareeba and to that extent renewed their past association;
(3)The trivial amount ($50 each) which Ms Deemal-Hall agreed to pay to the assailants was a sum that a jury might regard as apt for an assault but not as a fee for causing grievous bodily harm to the deceased, still less his death;
(4)Ms Deemal-Hall's reported instructions were that the assault was to amount to a "touch-up", "getting into" or "fixing up" the deceased or giving him a "hiding";
(5)The fact that the assault began with punches consistent with what arguably had originally been intended;
(6)The fact that the assailants did not themselves know the deceased and had no apparent reason to display personal violence or hostility towards him except to fulfil their $50 bargain with Ms Deemal-Hall;
(7)The fact that the deceased was aware of Ms Deemal-Hall's part in the assault; that she had been seen with him in the days before his death; and that he was left in a public place where he would easily and quickly be found; and
(8)The fact that there was no evidence before the attack to suggest that Ms Deemal-Hall was aware of the particular strengths of the assailants, of their propensities, of their footwear or of their possession of a pick handle or their intention to use it against the deceased.
So far as the male appellants are concerned, it is necessary to decide which version of the facts is correct, if either, given that the jury were confined, essentially, to versions respectively given to police by those who had participated to some degree in the assaults (including Mr Bowen), necessitating considerable care in the evaluation of what each of them had to say, exculpating themselves and inculpating the others. Making that evaluation is quintessentially a jury function. It is a function to be performed by the application of the correct legal standard expressed in terms of probability and not possibility. For the final appellate court to perform that function for the first time is for it to assume a role that belongs properly to the appellants' jury.
If one asks whether it was a possibility (even a "real possibility") that the events which Ms Deemal-Hall counselled would descend to the level of violence that they did, the answer, almost inevitably, is in the affirmative. Virtually anything is possible. Where violence gets out of hand, it is really possible. Yet it might not have been regarded as "a probable consequence" – particularly because of the considerations I have identified. At least, this was a view properly open to the jury. Similarly, with the male appellants, it all depended on whose version of events was accepted, given the imperfections of the versions that were available.
It cannot be the case that the "consequence" in question is judged with the benefit of hindsight after it has occurred. To judge whether it is "a probable consequence", it is necessary to place oneself before the events unfold. This means that the actual happenings cannot decide the case in judging the probabilities prospectively. The issue is one of foresight; but it is foresight objectively and not subjectively assessed.
Could it then be said that, before the attacks on the deceased unfolded, they were "a probable consequence" of the appellants' unlawful purpose? Having regard to the considerations that I have mentioned, a jury, properly instructed on the provisions of the Code, might answer that question in the negative. They might regard what happened as a series of events that got out of control and went far beyond the purposes for which the second appellant had engaged the male appellants and beyond the objects that she had counselled them to perform.
The more complex the evidence, the more important it was to have the correct legal standard applied to it. I realise that, by close analysis of the record and logical reasoning, it is possible to identify ways in which the verdicts entered against the appellants might be sustained. The unedifying nature of the attack on the deceased, and the cowardly way he was left to drown in his own blood, scarcely enliven sympathy for any of the appellants. However, they have been convicted of the most serious crime in the Code. They have been sentenced to the highest punishment provided by that law[148]. They were entitled, in such a case, to a very high degree of accuracy in the conduct of their trials. For such serious outcomes and to sustain such extended and multiple deprivations of liberty, it is normal in our system of justice to rely on the verdicts of juries accurately instructed. Their verdicts are preferable to the logical reasoning of judges – even of the judges of the nation's final court. The jury's collective wisdom might not follow the linear logic of the justifications of the guilty verdicts found in the joint reasons[149].
[148]Cf Anderson v The Queen [1972] AC 100 at 108.
[149]Joint reasons at [86]-[109]. See Stevens (2005) 80 ALJR 91 at 109-110 [80]-[82]; 222 ALR 40 at 62.
Unknown mode of jury reasoning:This conclusion is reinforced by a reflection on the holding of this Court in Domican v The Queen[150]. In that case, the evidence against the accused, looked at globally and with all of its component parts, was, as in these proceedings, very powerful. However, there was a defect in the judge's instruction to the jury on one issue (identification). This Court pointed out that, in the absence of a special verdict, appellate courts are not ordinarily able to conclude how the jury reasoned to their verdict. Because the reasoning might have been affected by the defective direction (if the jury chose to reason on the basis of the evidence of identification), the verdict and the conviction that depended upon it could not stand.
[150](1992) 173 CLR 555 at 566.
Verdicts that follow legal misdirection might not have been contaminated by the error. But the appellate court can never know. As in Domican, so here. If the jury in these proceedings reasoned to their verdicts by way of a conclusion that the appellants were guilty because there was a (real) possibility or chance that the death of the deceased would be a consequence of the appellants' conduct, that would render the misdirection critical. Other later rationalisations by judges, to support the verdicts on other lines of reasoning, would not then truly sustain the verdicts actually returned. The appellants would have lost their right to a jury trial according to law. That, without more, would be a miscarriage of justice of a serious kind suggesting the need for a retrial.
The exceptional basis of liability:There is an additional consideration. It derives from the fact that liability under ss 8 and 9 of the Code derogates from the normal requirement of Australian criminal law that a person is only criminally responsible for the acts that he or she intends. Here, the Code language adopts an objective standard. By a legal fiction, it imposes a kind of vicarious criminal responsibility, judged by an objective standard. It does this whatever the actual intention and purpose of the accused might have been[151].
[151]Cf Lane 474 US 438 at 462 (1986) per Brennan J.
Whilst the Code must be applied according to its terms, where those terms have been mis-stated, the exceptional character of the resulting criminal liability (and its extremely serious consequences in these proceedings) renders the mistake a very significant one. Where the mis-statement has been made with judicial authority to the decision-maker (here the jury), the potential for occasioning a miscarriage of justice is virtually inescapable.
Practical and policy considerations: In addition to the foregoing reasons for requiring a retrial, there are a number of practical or policy considerations that reinforce my conclusion that it would be inappropriate to apply the 'proviso' in the present circumstances. They include:
(1)Applying the 'proviso' effectively reduces this Court's extensive observations on the mis-statement of the law in Queensland (and the error of the judicial directions to the jury in the present case) to something bordering on an advisory opinion. True, it is not strictly so, for it is necessary, in applying the 'proviso', to conclude that "the point … raised by the appeal might be decided in favour of the appellant". But if it "might", that is enough. The Court is then directed to the issues of "substantial miscarriage of justice" with which the 'proviso' is concerned. In a court of error, such as this, where a serious mistake of law is revealed, there is a strong reason of principle why such mistakes should ordinarily be marked by the provision of relief and an order for a retrial. Indeed, this is the primary instruction of the Code itself ("shall allow the appeal").
(2)The foregoing is especially so where the mistake has involved a misdescription to the jury of the ingredients of the offences charged against the appellants[152]. Misdirections of such a kind are more serious than others. In Nudd v The Queen[153], another Queensland appeal, which concerned alleged incompetence of the appellant's legal representatives, I agreed to the application of the 'proviso', although analysis of the case, by reference to the fairness of the trial, occasioned me disquiet. In that case, however, it was found that "the trial judge instructed the jury on the applicable law with complete accuracy"[154]. Such was not done here. The increasing insistence of appellate courts upon the accurate explanation to juries of the central ingredients of the offence(s) charged is, in my opinion, a reason for the greater caution in the intermediate courts in the application of the 'proviso' in recent years[155]. Nothing said by the Court in Weiss suggests a reversal of that caution. It really speaks for itself. If the decision-maker in the trial (the jury) is misled as to its essential function and provided with an incorrect statement of the applicable legal components of the offence, the postulate of a trial according to law is not fulfilled. No amount of appellate reasoning can then replace that normal entitlement belonging to all persons accused of serious crimes[156]. The 'proviso' assumes that the essential postulate has been fulfilled. Here it has not been.
(3)Finally, there is a pragmatic concern. If appellate courts (and especially this Court) do not follow the finding of a basic legal error in criminal trials with the ordinary provision of relief proper to that finding, the likelihood is that appeals on such points will not be funded. They will then not be brought to a hearing. This will be so because the deemed "merits" seem likely to invite the 'proviso' and swamp the practical significance of establishing the error in question. Bringing appeals will be viewed as a waste of time, effort and resources. Important legal principles are often established in unsympathetic circumstances. Such are the circumstances here. But given the sentences the appellants were facing if convicted, the correct understanding by the jury of the considerations necessary to arrive at a guilty verdict was all the more important.
[152]See KBT v The Queen (1997) 191 CLR 417 at 433-434; Krakouer v The Queen (1998) 194 CLR 202 at 216-217 [36]-[37], 227 [76]; King v The Queen (2003) 215 CLR 150 at 182 [103]; Arulthilakan v The Queen (2003) 78 ALJR 257 at 269 [61]; 203 ALR 259 at 274.
[153](2006) 80 ALJR 614; 225 ALR 161.
[154](2006) 80 ALJR 614 at 637 [110]; 225 ALR 161 at 190.
[155]R v Whittaker (1993) 68 A Crim R 476 at 484; Gilbert v The Queen (2000) 201 CLR 414 at 438 [86] per Callinan J; Doggett v The Queen (2001) 208 CLR 343 at 384-385 [153].
[156]Antoun (2006) 80 ALJR 497 at 503 [23]; 224 ALR 51 at 57.
Orders
I would allow the appeals. I would set aside the judgment of the Court of Appeal of the Supreme Court of Queensland. In its place, I would order that the convictions of the appellants be quashed and that new trials be had.