R v Stephen; R v Tadrosse (No 4)

[2025] NSWSC 824

Supreme Court

New South Wales

Medium Neutral Citation:

R v Stephen; R v Tadrosse (No 4) [2025] NSWSC 824

Hearing dates:

22 July 2025

Date of orders:

25 July 2025

Decision date:

25 July 2025

Jurisdiction:

Common Law

Before:

N Adams J

Decision:

It is open to the Crown to run its case on extended joint criminal enterprise principles in this matter.

Catchwords:

EVIDENCE – pre-trial ruling – joint trial – charges of murder and assault with intent to rob, with wounding and armed – whether the Crown can advance a case of extended joint enterprise – where the Crown cannot prove which of the two participants in the joint criminal enterprise is the principal offender

Legislation Cited:

Crimes Act 1900 (NSW), ss 18, 98

Criminal Code 1899 (Qld), s 7(1)(a), (d)

Criminal Law Consolidation Act 1935 (SA), s 11, s 12A

Cases Cited:

Baker v Smith [2021] QCA 66

Batak v The King [2024] NSWCCA 66

Blundell v R [2019] NSWCCA 3

Carbone v The Queen [2020] NSWCCA 318

Carlyle-Watson v The Queen [2019] NSWCCA 226

CCIG Investments Pty Ltd v Schokman (2023) 278 CLR 165; [2023] HCA 21

Clayton v The Queen [2006] HCA 58; 168 A Crim R 174

Crane v R [2024] NSWCCA 112

Fan v The King [2024] NSWCCA 114

Gillard v The Queen (2003) 219 CLR 1; [2003] HCA 64

IL v The Queen (2017) 262 CLR 268; [2017] HCA 27

Jaghbir v R [2023] NSWCCA 175

Johns v The Queen (1980) 143 CLR 108; [1980] HCA 3

McAuliffe v The Queen (1995) 183 CLR 108; [1995] HCA 37

Miller v The Queen (2016) 259 CLR 380; [2016] HCA 30

Mitchell v The Queen (2023) 276 CLR 299; [2023] HCA 5

Mohan v The Queen [1967] 2 AC 187

Osland v The Queen (1998) 197 CLR 316; [1998] HCA 75

R v Batak [2025] HCA 18; 99 ALJR 820

R v DJD [2023] NSWSC 222

R v Jogee [2017] AC 387; [2016] UKSC 8

R v Lowery & King (No 2) [1972] VR 560

R v Phan (2001) 53 NSWLR 480; [2001] NSWCCA 29

R v Walsh [2002] VSCA 98; 131 A Crim R 299

Robertson v R [2024] NSWCCA 99

Category:

Procedural rulings

Parties:

Rex (Crown)
Danny Stephen (Accused)
Tony Phillip Tadrosse (Accused)

Representation:

Counsel:
Mr E Balodis (Crown)
Mr K Prince (Accused Stephen)
Ms M Avenell SC (Accused Tadrosse)

Solicitors:
Office of the Director of Public Prosecutions (Crown)
City Group Legal (Accused Stephen)
Crimcorp Defence Lawyers (Accused Tadrosse)

File Number(s):

2023/00116714
2023/00116875

Publication restriction:

Nil.

JUDGMENT

1. Tony Tadrosse and Danny Stephen both stand charged with two counts:

Count 1: The murder of Kalim Saliba (contrary to s 18(1)(a) of the Crimes Act 1900 (NSW)): and

Count 2: The assault with intent to rob of Shahidy Saliba with wounding whilst being armed (contrary to s 98 of the Crimes Act 1900 (NSW)).

2. Their trial is listed to commence before me and a jury on Monday 28 July 2025.

3. This judgment concerns a pre-trial application made on behalf of both accused that I rule that it is not open to the Crown, as a matter of law, to bring its case against both accused on the basis of extended joint criminal enterprise. Senior counsel for Mr Tadrosse based this application on the decision of Gordon, Edelman and Steward JJ in Mitchell v The Queen (2023) 276 CLR 299; [2023] HCA 5 (“Mitchell”) at [61]. Counsel for Mr Stephen joined that application. The Crown’s position is that there is nothing in Mitchell that precludes the Crown from bringing its case on principles of extended joint criminal enterprise on the facts in this trial.

4. The point is a simple but important one: can the Crown rely on principles of extended joint criminal enterprise when the Crown cannot specify which of the two parties to the agreement did the physical act which constitutes the charge/crime(s) which fell outside of the agreement but was foreseen as a reasonable possibility?

5. It was common ground that prior to the decision in Mitchell the Crown could (and did) bring cases on this basis. It was also common ground that since Mitchell the question raised on behalf of the accused in this application has not arisen for consideration by the High Court, the Court of Criminal Appeal or any intermediate appellate court in Australia. Senior counsel for the accused accepted that, apart from the decision in Mitchell at [61] (which did not expressly address the point raised in this application), there is no published decision that supports her application nor is there any published decision since Mitchell which supports the Crown position. There is simply an absence of authority on this issue.

6. It may seem surprising that there is no relevant authority on an area of the law as frequently considered by the High Court as joint criminal enterprise (both straightforward and extended). Prior to the decision in Mitchell in 2023, the relevant principles were considered by the High Court in, inter alia, Johns v The Queen (1980) 143 CLR 108; [1980] HCA 3 (“Johns”), McAuliffe v The Queen (1995) 183 CLR 108; [1995] HCA 37 (“McAuliffe”), Osland v The Queen (1998) 197 CLR 316; [1998] HCA 75 (“Osland”), Gillard v The Queen (2003) 219 CLR 1; [2003] HCA 64 (“Gillard”), Clayton v The Queen [2006] HCA 58; 168 A Crim R 174 (“Clayton”), Miller v The Queen (2016) 259 CLR 380; [2016] HCA 30 (“Miller”) and IL v The Queen (2017) 262 CLR 268; [2017] HCA 27 (“IL”).

7. Despite the fact that the High Court has not always used consistent language in this area, prior to Mitchell in 2023, the High Court had consistently described both straightforward and extended joint criminal enterprise as relying on principles of principal rather than derivative liability.

8. In Mitchell, three of the seven judges of the High Court (Gordon, Edelman, and Steward JJ) described extended joint criminal enterprise for the first time as being a form of “derivative” liability. The other four judges (Kiefel CJ, Gageler, Gleeson and Jagot JJ) did not describe extended joint criminal enterprise as a “derivative” form of liability; their judgments are silent on this issue.

9. The question raised on this application concerns the extent, if any, to which the description of extended joint criminal enterprise as “derivative” for the first time in Mitchell has fundamentally changed the way the Crown can bring a case where the identity of the principal offender, although confined to a person within a specified group, cannot be established beyond reasonable doubt.

10. I propose to first outline the Crown case, but only to the extent necessary to explain the issue that has arisen on the facts of this case. The Crown Case Statement is very detailed, and I have only extracted brief portions of it. I then propose to set out the relevant authorities in some detail before turning to address the respective arguments.

Background facts

11. On 29 April 2020, at about 12:15am, 86-year-old Kalim Saliba and his 83-year-old wife Shahidy Saliba were watching television in their home in Cherrybrook when, on the Crown case, one or two persons entered their home to assault them with the intention to rob them. The glass panel of their front door was smashed to gain entry. Mr and Mrs Saliba were not upstairs asleep in their bedroom as might have been suspected at that late hour; they had stayed up watching television and were instead in the rumpus room. They came across a male intruder who was wearing a balaclava (consistent with the contemplation that he would be confronting one or both of the residents). He was at that time armed with what Mrs Saliba described as a 40cm wooden stick with a ball-shaped end.

12. The intruder said, “[w]here’s the money, where’s the money” and when Mr and Mrs Saliba yelled at him that they did not have any money he beat them both with that weapon. Mr Saliba died, Mrs Saliba was rendered unconscious and the intruder left. Mrs Saliba was able to call for help. When police arrived, she told them that whilst the intruder was assaulting her and her husband, she could hear another intruder upstairs. At no time did Mrs Saliba ever see two intruders at the same time.

13. At the hearing, the Crown submitted that while the attempted robbery was taking place the getaway car was running. The following exchange took place: [1]

“HER HONOUR: … I understood that was relied upon to say there were two in the house, but was the car running?

BALODIS: The car was running.

HER HONOUR: The car was running? Okay.”

1. Tcpt, 22 July 2025, p 27(22-27).

14. Although this fact does not appear in the Crown Case Statement, it seemed to be common ground that there is evidence to be adduced at the trial that the car was running, implying that there may have been a third offender. Another neighbour, Isolda Ferreiro, describes seeing only one person leave the premises.

15. There is no dispute that Mr and Mrs Saliba confronted a masked intruder in their home who then assaulted them both with a weapon. The dispute is as to the identity of the assailant and those who acted with him.

16. The Crown case is that Mr Tadrosse and Mr Stephen reached an agreement prior to 29 April 2020 to enter the premises at night, at a time when the two residents would be home and assault them with intent to rob them of cash which they knew would be on the premises. In the Crown Case Statement dated 7 July 2025, the Crown puts its case this way at [6]:

“The Crown contends that each accused is guilty of each of counts 1 and 2 by the doctrine of extended joint criminal enterprise. Each accused was a party to a joint criminal enterprise, in that they entered into an agreement to assault the occupants of the house (that is, to cause them to apprehend immediate and unlawful violence) with intent to rob them of money or other valuable property, while being armed with an offensive weapon or instrument.

In relation to Count 1, each accused foresaw the possibility that one of them, in carrying out the joint criminal enterprise to assault with intent to rob while armed, might intentionally cause death or grievous bodily harm to an occupant of the house.

In relation to Count 2, each accused foresaw the possibility that one of them, in carrying out the joint criminal enterprise to assault with intent to rob while armed, might wound an occupant of the house.”

17. The Crown case is circumstantial as to the identification of the offenders based on, inter alia, the following:

1. Both of the accused are related to the Salibas; Mr Tadrosse is their nephew, and Mr Stephen is their great-nephew;

2. Both men were in financial difficulty at the time, and thus had a motive;

3. The Salibas used to run a successful fruit and vegetable business and Mr Stephen (at least) knew that they had cash at their home. On the Crown case, both men had knowledge that there would be cash at the premises;

4. The Salibas did in fact have a safe at their house that contained $52,030. The Crown case is that only someone who had knowledge of the Salibas would know that they kept money at their home;

5. A combination of mobile telephone use and CCTV evidence will be relied upon to establish that both Mr Stephen and Mr Tadrosse were either together or in close contact with each other in the lead up to and including the time of the offences;

6. Mr Stephen was associated with an Audi vehicle similar to that seen doing reconnaissance of the Salibas’ house that day. The Crown relies on CCTV footage to establish that this was the only vehicle that drove in and out of Cherrybrook that night (in reliance on CCTV footage etc);

7. Mr Tadrosse, who resided in Potts Point at the time, travelled by train to Thornleigh (near Cherrybrook) in the afternoon before the attempted robbery of the Salibas (his Opal card was used at these stations and his telephone(s) was detected following that route);

8. Both men used covert telephones to communicate with each other. Those phones place them in the Cherrybrook area at the relevant time;

9. After the attempted robbery, Mr Stephen drove Mr Tadrosse home to Potts Point. The Crown relies on CCTV footage of his car, toll usage and telephone activity to establish this;

10. After Mr Tadrosse arrived home, there was telephone contact between both men. After that time, Mr Stephen transferred $150 to Mr Tadrosse’s bank account who then caught an Uber in his own name back to Pennant Hills and then home again at a cost of $176.65;

11. When Mr Tadrosse got home, he again spoke to Mr Stephen at 3:25am;

12. At 4:44am, Mr Tadrosse telephoned the COVID hotline and stated that he had just woken up, that he had not been out except to buy groceries and that he had been tired and had symptoms all day. The Crown case is that in doing so he set up a false alibi; and

13. Mr Stephen was subsequently questioned by police about whether he had been driving the Audi that night and lied to police that he had not. The Crown relies upon this as consciousness of guilt.

18. The application before me was that the Crown not be permitted to bring its case in the way described at [6] of the Crown Case Statement. Although the joint criminal enterprise is described in that paragraph as only being between the two accused, during the hearing of this application it emerged that, on the Crown case, there may be a third participant in the enterprise. Although there is nothing in the Crown case to suggest the involvement of anybody else, the Crown pointed to three aspects of the evidence which raises the possibility of an unknown third offender. Those three aspects of the evidence are:

1. Mrs Saliba describes hearing somebody moving upstairs whilst she was being assaulted;

2. A neighbour only saw one person leave the premises;

3. The evidence that the Audi was left running suggests that there may have been somebody in the car whilst the attempted robbery was taking place (although not necessarily).

19. I shall proceed to determine the application based on the terms of [6] of the Crown Case Statement and will return to the significance, if any, of a possible third offender later in these reasons.

20. The Crown has evidence capable of establishing, by the drawing of inferences, that: the two men planned the robbery; that they were in the only vehicle which entered and exited or drove in and out of Cherrybrook at a time when the offences took place, and that one of them was the person who physically attacked the Salibas. Despite this, the Crown accepts that it cannot prove beyond reasonable doubt which of Mr Tadrosse or Mr Stephen was the person who physically assaulted Mr and Mrs Saliba.

Principles of complicity

21. The issue raised on this application is the extent to which the elements the Crown has to prove in a prosecution based on principles of extended joint criminal enterprise have changed since it was characterised as a form of derivative rather than primary liability. This requires an understanding of these two respective forms of liability. Apart from the issue the subject of this application, the general principles of complicity were accepted by both the Crown and the accused.

22. It is well established that, by proper application of principles of criminal complicity, a person may be found criminally liable for the physical acts of another. The parties proceeded on the basis that in New South Wales common law principles of complicity provide for three different ways by which a person may be found criminally liable for the physical acts of another. They are:

1. Joint criminal enterprise;

2. Extended joint criminal enterprise; and

3. Accessorial liability.

23. These three principles overlap to some extent and are sometimes relied upon by the Crown in the alternative. Although the Crown does not rely on principles of accessorial liability in this trial, given that accessorial liability is a form of derivative liability, it was common ground that some of the principles derived from cases concerning accessorial liability may be of assistance in the determination of this application.

Accessorial liability

24. The common law principles of accessorial liability pre-date the common law principles of joint criminal enterprise by a significant margin; accessorial liability principles have been around for centuries whereas principles of joint criminal enterprise are more recent. I summarised the relevant principles in Blundell v R [2019] NSWCCA 3 at [91]-[108]. They have been more recently summarised by the Court of Criminal Appeal (Kirk JA, Wilson and Irace JJ) in Batak v The King [2024] NSWCCA 66 at [51]-[70] (“Batak”).

25. The cases concerning accessorial liability use language such as “principals in the second degree”, “accessories before the fact” and “aiders and abettors”. These expressions all refer to a person who is liable to be sentenced as a principal for assisting that principal in some way. The principal offender is the person (or one of the persons) who physically does the act that amounts to the crime. There is usually no need to rely on any principle of complicity in order to prosecute such an offender (“P1”). A principal in the second degree is a person who assisted P1 either before the crime was committed or at the scene by intentionally aiding or giving encouragement to P1 (“P2”). I propose to adopt the descriptors “P1” and “P2” in these reasons for ease of reference.

26. For P2 to be guilty as an accessory either before the fact or at the scene the prosecution must prove beyond reasonable doubt:

1. That P1 committed the offence; and

2. That P2 knew of all of the facts and circumstances (both of a physical and mental nature) which make up that offence; and

3. That with that knowledge P2 intentionally provided assistance to P1 (either before or at the scene).

27. Thus, the first element in a case based on accessorial liability is to prove, beyond reasonable doubt, that P1 physically did the act amounting to a crime in order to prosecute other persons. It is for this reason that accessorial liability is described as derivative. It is a derivative form of liability because P2’s liability is wholly dependent on P1 being guilty. If the Crown cannot establish beyond reasonable doubt that P1 committed the offence, then none of the accessories can be convicted. It is this element of accessorial liability that the accused rely upon in the application before me.

28. I note that accessory after the fact is a separate offence and should not be confused with being an accessory before or at the fact. It does not rely on principles of derivative liability and does not arise in this case.

Joint criminal enterprise

29. As the Court of Criminal Appeal observed in Batak at [82], although the principles of joint criminal enterprise (“JCE”) can be traced to the decision in Johns, it is the unanimous decision of the High Court in McAuliffe that is generally regarded as being one of the classic statements of the law of criminal complicity with respect to issues of joint criminal enterprise and extended joint criminal enterprise (“EJCE”). Their Honours observed the following at 113-114 (citations omitted):

“The doctrine of common purpose applies where a venture is undertaken by more than one person acting in concert in pursuit of a common criminal design. Such a venture may be described as a joint criminal enterprise. Those terms – common purpose, common design, concert, joint criminal enterprise – are used more or less interchangeably to invoke the doctrine which provides a means, often an additional means, of establishing the complicity of a secondary party in the commission of a crime. The liability which attaches to the traditional classifications of accessory before the fact and principal in the second degree may be enough to establish the guilt of a secondary party: in the case of an accessory before the fact where that party counsels or procures the commission of the crime and in the case of a principal in the second degree where that party, being present at the scene, aids or abets its commission. But the complicity of a secondary party may also be established by reason of a common purpose shared with the principal offender or with that offender and others. Such a common purpose arises where a person reaches an understanding or arrangement amounting to an agreement between that person and another or others that they will commit a crime. The understanding or arrangement need not be express and may be inferred from all the circumstances. If one or other of the parties to the understanding or arrangement does, or they do between them, in accordance with the continuing understanding or arrangement, all those things which are necessary to constitute the crime, they are all equally guilty of the crime regardless of the part played by each in its commission.

Not only that, but each of the parties to the arrangement or understanding is guilty of any other crime falling within the scope of the common purpose which is committed in carrying out that purpose. Initially the test of what fell within the scope of the common purpose was determined objectively so that liability was imposed for other crimes committed as a consequence of the commission of the crime which was the primary object of the criminal venture, whether or not those other crimes were contemplated by the parties to that venture. However, in accordance with the emphasis which the law now places upon the actual state of mind of an accused person, the test has become a subjective one and the scope of the common purpose is to be determined by what was contemplated by the parties sharing that purpose.”

30. The High Court referred to three different forms of JCE in this passage which I will return to below. Significantly, a basis for liability in a JCE case of being guilty for what was contemplated emerged: EJCE.

31. In Gillard, the High Court endorsed the principle of EJCE followed by Clayton. In fact, the appellants in Clayton invited the High Court to reconsider McAuliffe and Gillard and find that liability arising from an agreement should be limited to JCE only (and not EJCE). They were unsuccessful six to one.

32. In 2016, the joint decision of the United Kingdom Supreme Court and the Privy Council in R v Jogee [2017] AC 387; [2016] UKSC 8 (“Jogee”) held that the doctrine of EJCE is not part of the common law of the UK. This led to another challenge to the doctrine of EJCE in Miller.

33. In Miller, the appellants again unsuccessfully invited the High Court to overrule the doctrine of EJCE as stated in McAuliffe. Six of the seven judges in Miller declined to follow the same approach as Jogee. Rather, it was held that if any change to the law is to be made, it should be made by the Parliament: Miller at [41]. The dissenting judge was Gageler J. In Miller, French CJ, Kiefel, Bell, Nettle and Gordon JJ, by reference to McAuliffe, explained the three types of joint criminal enterprise as follows at [4]:

“The law, as stated in McAuliffe, is that a joint criminal enterprise comes into being when two or more persons agree to commit a crime. The existence of the agreement need not be express and may be an inference from the parties' conduct. If the crime that is the object of the enterprise is committed while the agreement remains on foot, all the parties to the agreement are equally guilty, regardless of the part that each has played in the conduct that constitutes the actus reus. Each party is also guilty of any other crime (the incidental crime) committed by a co-venturer that is within the scope of the agreement (joint criminal enterprise liability). An incidental crime is within the scope of the agreement if the parties contemplate its commission as a possible incident of the execution of their agreement. Moreover, a party to a joint criminal enterprise who foresees, but does not agree to, the commission of the incidental crime in the course of carrying out the agreement and who, with that awareness, continues to participate in the enterprise is liable for the incidental offence (‘extended joint criminal enterprise’ liability).”

34. It should be noted that the transcript of the submissions before the High Court in Miller [2] reflect a submission by senior counsel for one of the applicants (at 3930) that:

“In respect of joint criminal enterprise, our position is that it is impossible to rationalise that as anything other than derivative. The reason is it is a different crime and the other reason is you have to prove that the other crime was committed by the individual.”

2. [2016] HCATrans 107.

35. There is no reference in the decision of Miller to that submission.

36. In IL, Kiefel CJ, Keane and Edelman JJ noted the difference between primary and derivative liability (at [30]) as follows:

“… the liability of persons as accessories before the fact to murder (ie persons not present at the commission of the crime) was a form of derivative liability. So too was the liability derivative for persons who were ‘merely present, encouraging but not participating physically, or whose acts were not a substantial cause of death’. But where two or more persons act in concert then any liability is primary. The acts of one are attributed to the others because they reached an understanding or arrangement that together they would commit a crime and the acts were performed in furtherance of that understanding or arrangement.”

37. Further, in IL Gageler J observed (at [107]):

“For completeness, I note that Osland concerned only the doctrine of joint criminal enterprise and not the doctrine of extended joint criminal enterprise recently affirmed by majority in Miller. This case too concerns only joint criminal enterprise. Whether criminal responsibility attributed by operation of the doctrine of extended joint criminal enterprise is primary or derivative and how, if at all, the doctrine of extended joint criminal enterprise might intersect with constructive murder are questions which do not now arise for consideration.” (Emphasis added.)

Elements of JCE/EJCE

38. It is helpful to set out the elements the Crown has to prove beyond reasonable doubt in a case of JCE/EJCE derived from the decisions I have summarised above.

39. The parties agree that there are three recognised categories of JCE and EJCE:

1. The participants agreed to commit a crime, and that crime is committed by one or more of the participants (whilst the agreement is still on foot) and each person did an act in furtherance of that agreement. In a JCE, all the parties to the agreement are equally guilty: McAuliffe at 114; Osland at [27], [33] (Gaudron and Gummow JJ), [72]-[73], [93] (McHugh J, Kirby J agreeing at [174], Callinan J agreeing at [257]); Gillard at [109]-[110] (Hayne J); Clayton at [14] (Gleeson CJ, Gummow, Hayne, Callinan, Heydon and Crennan JJ); Miller at [4] (French CJ, Kiefel, Bell, Nettle and Gordon JJ); IL at [2], [29]-[30] (Kiefel CJ, Keane and Edelman JJ), [61], [66] (Bell and Nettle JJ); Mitchell at [13]-[14] (Kiefel CJ), [54]-[55], [61] (Gordon, Edelman and Stewart JJ).

2. The crime committed by one or more of the participants is within the scope of the agreement because it is a possible incident of carrying out the offence which is the subject of the joint criminal enterprise (the incidental crime): Johns at 130-131 (Mason, Murphy and Wilson JJ); McAuliffe at 114-115; Gillard at [111] (Hayne J); Clayton at [14] (Gleeson CJ, Gummow, Hayne, Callinan, Heydon and Crennan JJ); Miller at [4], [10] (French CJ, Kiefel, Bell, Nettle and Gordon JJ); IL at [2], [26], [29]-[30] (Kiefel CJ, Keane and Edelman JJ), [62], [66] (Bell and Nettle JJ); Mitchell at [13]-[14] (Kiefel CJ), [33] (Gageler, Gleeson and Jagot JJ), [52], [54]-[55], [61] (Gordon, Edelman and Stewart JJ).

3. The crime committed was outside the scope of the agreement but the accused foresaw or contemplated that the incidental crime might be committed in the course of carrying out the agreement and continued to participate with that awareness: McAuliffe at 115, 117-118; Gillard at [112], [118]-[119] (Hayne J); Clayton at [14] (Gleeson CJ, Gummow, Hayne, Callinan, Heydon and Crennan JJ); Miller at [4], [10] (French CJ, Kiefel, Bell, Nettle and Gordon JJ); IL at [34] (Kiefel CJ, Keane and Edelman JJ), [63] (Bell and Nettle JJ); Mitchell at [15] (Kiefel CJ), [52], [56]-[57], [61] (Gordon, Edelman and Stewart JJ).

40. The decisions of Miller and Mitchell confirm that both (a) and (b) are forms of JCE as what occurred was within the scope of the agreement whereas category (c) is a form of EJCE.

41. When the Crown relies on (a), it must prove beyond reasonable doubt (in addition to the commission of the crime) both:

1. The existence of a joint criminal enterprise to commit the specified crime; and

2. The participation in that specified crime by the accused.

42. This form of complicity does not arise in this matter.

43. When the Crown relies on (b), it must prove beyond reasonable doubt (in addition to the commission of the crime):

1. The existence of a joint criminal enterprise to commit the specified crime;

2. The participation in that specified crime by the accused; and

3. That the commission of the incidental crime was within the scope of that agreement.

44. When the Crown relies upon (c), it must prove beyond reasonable doubt (in addition to the commission of the crime):

1. The existence of a joint criminal enterprise to commit the specified crime;

2. The participation in that specified crime by the accused; and

3. That although the commission of the incidental crime was not within the scope of that agreement, it was contemplated as a reasonable possibility it might be committed.

45. A question arises as to the extent to which the decision in Mitchell has changed the elements of EJCE. Before turning to consider the decision in Mitchell, it is relevant to mention the decision in Osland as it is an important case highlighting the significant difference between JCE/EJCE (primary liability) and accessorial liability (derivative liability).

Osland v The Queen

46. The significance of whether liability is to be described as primary, or derivative is best illustrated by the decision in Osland. That case concerned the question of whether there had been inconsistent verdicts and required the High Court to explain the differences between primary and derivative liability.

47. Mrs Osland and her son were charged with murdering Frank Osland, the husband and father of the two accused respectively. He had been abusive to them for years. The Crown case was that Mrs Osland mixed sedatives into the deceased’s dinner and, while he was sleeping, her son fatally struck him with an iron pipe in the presence of the appellant.

48. The appellant and her son were tried together and neither disputed that they had killed the deceased. Rather, each of them relied on provocation and self-defence on the basis of years of violent abuse. Mrs Osland was convicted of murder, but the jury was hung with respect to her son. The son was later re-tried and ultimately acquitted. The case had not been brought as one of accessorial liability; both accused were prosecuted on the basis of straightforward JCE to murder the deceased.

49. The main issue for the High Court was whether the verdicts between the appellant and her son were logically inconsistent. In his judgment dismissing the appeal, McHugh J looked at the nature of different types of criminal liability and complicity as follows (at 341-343) (footnotes omitted):

“At common law, a person who commits the acts which form the whole or part of the actus reus of the crime is known as a ‘principal in the first degree’. There can be more than one principal in the first degree. However, a person may incur criminal liability not only for his or her own acts that constitute the whole or part of the actus reus of a crime but also for the acts of others that do so. The liability may be primary or derivative.

…

Those who aided the commission of a crime but were not present at the scene of the crime were regarded as accessories before the fact or principals in the third degree. Their liability was purely derivative and was dependent upon the guilt of the person who had been aided and abetted in committing the crime. Those who were merely present, encouraging but not participating physically, or whose acts were not a substantial cause of death, were regarded as principals in the second degree. They could only be convicted of the crime of which the principal offender was found guilty. If that person was not guilty, the principal in the second degree could not be guilty. Their liability was, accordingly, also derivative.

However, there is a third category where a person was not only present at the scene with the person who committed the acts alleged to constitute the crime but was there by reason of a pre-concert or agreement with that person to commit the crime. In that category, the liability of each person present as the result of the concert is not derivative but primary. He or she is a principal in the first degree. In that category each of the persons acting in concert is equally responsible for the acts of the other or others. The general principle was clearly stated in R v Lowery and King [No 2] by Smith J …”

50. In short, had the case been brought as one of accessorial liability, Mrs Osland could not have been guilty (as her liability derived from her son’s guilt).

Mitchell v The Queen

51. The special leave point in Mitchell was whether liability for murder under s 11 of the Criminal Law Consolidation Act 1935 (SA) (“the Act”) could be established by combining the common law doctrine of EJCE with constructive murder as provided for by s 12A of that Act. It should be noted that the language of s 12A is not the same as s 18(1) of the Crimes Act (NSW).

52. The appellants were alleged to have agreed to break and enter a drug house to steal cannabis. In doing so, one or more of the participants violently assaulted the deceased causing his death. The prosecution case was run on two bases (and not strictly as alternative cases) described in the judgment as “two pathways”.

53. One of the two “pathways” to murder was in reliance upon common law principles of EJCE. The prosecution alleged that the appellants were guilty of murder on the basis that they each foresaw the possibility that, in carrying out the agreement to break and enter and steal the cannabis, one of their co-venturers might attack the deceased with an intention to kill or cause grievous bodily harm.

54. The other pathway argued by the Crown was that the doctrine of EJCE operated together with s 12A of the Act to make each of the appellants guilty of murder under s 11 of the Act because each had foreseen that, in the course of carrying out their agreement, one of them might have perpetrated “an intentional act of violence”. That made the mental element for liability for murder: “an intentional act of violence” and the trial judge provided the example of hitting someone on the back of his leg with a stick. The jury was directed that for the second of the two pathways (EJCE constructive murder) the prosecution did not need to prove that the appellants contemplated that one of them might commit the act of violence with intention to kill or cause really serious bodily harm.

55. The Court of Appeal dismissed the appellants’ appeals against conviction.

56. The ratio of the High Court decision in Mitchell is that common law principles of EJCE and s 12A of the Act could not be relied upon in combination to create a new pathway to murder (s 11 of the Act). The result in Mitchell turned on a question of statutory construction. Despite the narrow issue for determination in Mitchell for the purposes of the law of New South Wales, it has effected a significant change to the common law principle of EJCE.

57. In Mitchell, Gordon, Edelman, and Steward JJ describe an EJCE as being derivative liability at [61] as follows:

“The rationale for extended joint criminal enterprise must be different from the rationale for joint criminal enterprise since extended joint criminal enterprise may render a participant in a criminal enterprise liable for a crime committed during that criminal enterprise which was foreseen but which fell outside the scope of the agreement. In this way, rather than being a principle of primary liability stemming from the agreement, extended joint criminal enterprise is a principle of derivative (or ‘secondary’) liability. Rather than being a principle that is dependent upon the scope of the agreement, it is a principle that depends upon whether the accused party to the agreement foresaw that an incidental crime might be committed, that is, a crime which was not ‘within the common purpose’ or agreement. Rather than having a rationale based upon the attribution of the authorised acts of another, its rationale is that a person is to be regarded as ‘intentionally assisting in the commission of a crime’ when that party ‘continues to participate in the venture without having agreed to, but foreseeing as a possibility, the act causing death’ and that act is foreseen as being coupled with the requisite intent. In other words, the liability of the accused is derived from, and dependent upon, the criminal liability of another for the foreseen crime that was not part of the agreement. Unlike joint criminal enterprise simpliciter, in extended joint criminal enterprise there is no attribution of the acts in respect of the incidental crime because the secondary participant did not authorise or agree to the commission of the incidental crime.” (Emphasis added.)

58. It was common ground that this was the first time EJCE has ever been described as derivative.

59. No other High Court judgments have considered Mitchell at [61], although I note that Edelman and Steward JJ have cited it in CCIG Investments Pty Ltd v Schokman (2023) 278 CLR 165; [2023] HCA 21 at [59] and Edelman J has cited it in R v Batak [2025] HCA 18; 99 ALJR 820 at [32].

Intermediate appellate decisions citing Mitchell

60. Prior to the decision in Mitchell, decisions of the Court of Criminal Appeal had proceeded on the basis that EJCE is a form of primary liability and accessorial liability is a form of derivative liability; see for example Carbone v The Queen [2020] NSWCCA 318 at [91] (Button J, Bathurst CJ and Wilson J agreeing) and Carlyle-Watson v The Queen [2019] NSWCCA 226 at [50]-[51] (Payne JA, Garling and Wright JJ agreeing).

61. Since the decision in Mitchell, this Court and the Court of Criminal Appeal have taken a different approach.

62. In R v DJD [2023] NSWSC 222, Wright J observed the following (at [71]):

“Extended joint criminal enterprise itself can be said to involve a form of constructive criminal responsibility since it gives rise only to secondary or derivative liability: Mitchell at [61]. In this regard, it is fundamentally different from joint criminal enterprise which gives rise to primary liability: Mitchell at [55].”

63. In Batak, the Court observed (at [92]):

“In Mitchell, Gordon, Edelman and Steward JJ stated that the nature of liability for EJCE differs from that for JCE. It does not involve agency or attribution of acts because the relevant acts were not part of what was agreed between the parties; liability is secondary and derivative (at [61]). The ‘liability of the accused is derived from, and dependent upon, the criminal liability of another for the foreseen crime that was not part of the agreement’ (ibid).”

64. In Robertson v R [2024] NSWCCA 99 (“Robertson”), Dhanji J observed (at [121]):

“In Mitchell v The King, Gordon, Edelman and Steward JJ (at [61]) expressed the view that, in contrast to the position which pertains in the case of a straight-forward joint criminal enterprise, liability for a crime on the basis of extended joint criminal enterprise is derivative. That is, the liability of the accused is derived from (and dependent on) the liability of the principal whose crime was foreseen as a possibility by the accused, even though the accused did not agree that the crime should be committed. Their Honours said that ‘in extended joint criminal enterprise there is no attribution of the acts in respect of the incidental crime because the secondary participant did not authorise or agree to the commission of the incidental crime’: Mitchell v The King at [61]. Following this approach, where it is sought to attach liability to an accused on the basis of an extended joint criminal enterprise it will be necessary for the Crown to prove the guilt of the principal (or at least the commission of the elements of the offence by the principal). This is consistent with the expression of six members of the High Court in Clayton v The Queen [2006] HCA 58; 81 ALJR 439 at [26]. It can also be understood as consistent with the basal principle that it is the agreement that operates to make the acts of the other participants to the joint criminal enterprise attributable to the accused. In the circumstances of the present case, this would require the Crown to prove the principal not only acted with the requisite intent but also that he did not believe his actions were necessary to protect himself or another.”

65. In Crane v R [2024] NSWCCA 112 (“Crane”), which was a case concerned with principles of accessorial liability, Dhanji J observed (at [85]):

“It was not suggested that derivative liability could flow to the applicant from the guilt of a principal based on extended joint criminal enterprise.”

66. In Fan v The King [2024] NSWCCA 114 (“Fan”), Button J (Bell CJ and McNaughton J agreeing) observed (at [59]-[60]):

“[59] Turning to the second leg of the Crown case, extended JCE, as I have said, the liability of the applicant pursuant to that doctrine has been clarified now to be derivative: see Batak v R at [92]; referring to Mitchell v The King at [61]. That meant that the jury in this trial needed to be satisfied beyond reasonable doubt of the guilt of the shooter, before it could return a verdict of guilty against the applicant.

[60] The derivative nature of the liability of the applicant pursuant to this doctrine, it patently not being based upon agreement, meant that the applicant could ‘get the benefit’ of any exculpatory matter pertaining to the shooter. So, for example (and leaving aside pathways to guilt based upon innocent agency and the like), if the shooter had been found to have been not criminally responsible due to a mental health impairment or cognitive impairment, or irrebuttably presumed to be incapable of committing an offence because he was under 10 years of age, then no liability could have been derived from that of the shooter and placed upon the shoulders of the applicant, and the latter would have been entitled to an acquittal as well.”

67. Thus, it can be seen that the Court of Criminal Appeal has proceeded on the basis that EJCE liability is derivative and not primary in Batak, Crane, Robertson, and Fan.

68. I have otherwise been unable to find any other intermediate appellate court or superior court decision outside NSW which has considered the question the subject of this application. That may be because NSW and South Australia are the only two states which rely on the common law for principles of complicity. Nor could any assistance be gained from English decisions given the decision in Jogee.

Accessorial liability cases relied upon by the Crown

69. In its submissions, the Crown relied on several cases concerning accessorial liability that are relevant to the question for consideration. In these decisions, the prosecution was permitted to run a case in which multiple accused were not distinguished as being either P1 or P2.

70. Mohan v The Queen [1967] 2 AC 187 concerned the accessorial liability for murder of two co-offenders, a father and son. Both accused had attacked the same man at the same time with similar weapons. The evidence was unclear as to which of the two accused inflicted the fatal wound. It also could not be inferred from the evidence that there was a pre-arranged plan for the attack. Principles of accessorial liability were relied upon.

71. The issue on appeal to the Privy Council (from the Supreme Court of Trinidad and Tobago) was whether each of the accused must be acquitted unless it was proved either that he inflicted the fatal wound or that they were both acting in pursuance of a pre-arranged plan. The Privy Council held that each accused could be guilty of murder as a principal in the second degree for aiding and abetting the commission of the offence. Their Lordships held (at 194):

“In this case one of the appellants struck the fatal blow and the other of them was present aiding and abetting him. In such a case the prosecution do not have to prove that the accused were acting in pursuance of a pre-arranged plan.”

72. In R v Lowery & King (No 2) [1972] VR 560, Smith J directed the jury as to the liability of two accused for the murder of a young girl. The Crown case alleged that the two accused were acting in concert as principals in the first degree or alternatively that one was the principal, and the other was the aider and abetter. The Crown did not specifically nominate which accused was the principal and which was the aider and abetter. In that context, Smith J directed the jury that (at 562):

“… even if for some reason you were not satisfied that there was an actual understanding or arrangement between the two accused that the girl should be killed, nevertheless you should at least be satisfied by the evidence admissible against each of the accused that the girl must have been intentionally killed by the conscious, voluntary act of one of them, no matter which, aided and abetted by the other.”

73. R v Phan (2001) 53 NSWLR 480; [2001] NSWCCA 29 (“Phan”) again concerned a case of accessorial liability where the Crown could not prove which person committed the relevant act. The deceased had been shot three times while sitting in a car with the accused and another person. The Crown case was that either the accused fired the shots or alternatively that he was an aider and abettor to the other person and liable as a principal in the second degree. On appeal it was held that the instructions given by the trial judge were inadequate in circumstances where the appellant asserted that his presence at the scene was accidental, that he was unconscious at the time of the shooting, and that he did not provide any encouragement or assistance to the shooter. A new trial was ordered.

74. Wood CJ at CL described the Crown case in this way (at [65]):

“… it was a case, if anything, of common enterprise in which the parties to it were each active participants and hence liable as principals in the first degree, or one where the Crown being unable to prove which of several persons, present at the scene of a crime, committed the particular act giving rise to the offence, nevertheless was able to establish that each was assisting the other, in which event it was properly left as one of principal and accessory: Mohan v The Queen (1967) 2 AC 187 and R v Clough (1992) 28 NSWLR 396.” (Emphasis added.)

75. Smart AJ described the Crown case in this way (citing Mohan v The Queen) (at [90]):

“Three shots were fired. Either one of the men fired all three shots or each man fired one or more shots. At least one of the shots killed the deceased. If the jury was not satisfied that the accused fired one of the shots that did not matter so long as they were satisfied that both the men were present, with one or both of them firing the shots and the other aiding and abetting. As Hunt J pointed out in R v Stokes & Difford (1990) 51 A Crim R 25 at 35, even if the Crown had been unable to establish which one of the two co-offenders was responsible for inflicting the injury it would still have been a classic case of each of them aiding and abetting the other: Mohan v The Queen [1967] 2 AC 187 at 195. In Mohan the evidence established that the two co-offenders were present and attacked the deceased. The Crown could not say who struck the fatal blow.” (Emphasis added.)

76. A similar approach was taken in R v Walsh [2002] VSCA 98; 131 A Crim R 299. Phillips and Buchanan JJA (Ormiston JA agreeing) considered a case in which the Crown put its case in the alternative. At [45]-[46], their Honours observed:

“[45] The clearest example of the jury's being permitted in such cases to arrive at the result by different routes is to be found where the Crown puts its case in the alternative, being unable to choose between them. It may be alleged that the victim was killed by the accused either personally or by an aide, the evidence leaving it unclear which, and in such a case it is enough if the jury agree that it was one or the other; beyond that, agreement is not required.

[46] That has been so, it seems, since Swindall v Osborne (1864) 2 Car & K 230 in which the victim was killed by one of two carts, the driver of one inciting the driver of the other to kill the deceased. The conviction of one driver was upheld because, if he was not the principal, he was an accessory, having incited the other.”

77. Similarly, in Baker v Smith [2021] QCA 66 (“Baker v Smith”), the Queensland Court of Appeal discussed the alternatives of principal or accessorial liability under ss 7(1)(a) and (d) of the Criminal Code 1899 (Qld), albeit in the context of offences of land clearing. It held at [19]:

“It is not uncommon for convictions to be based on proof of one or more possibilities, with satisfaction beyond reasonable doubt that there was no reasonable hypothesis of any other possibility. For example, where two or more offenders jointly attack and kill a victim, it will often not be possible to determine which of them delivered the lethal blow. As a matter of fact one will have done so, and as a matter of law, he is the actor and liable through s 7(1)(a) of the Code and the others are aiders and liable under s 7(1)(b). However, if the jury is satisfied beyond reasonable doubt that any particular offender either delivered the fatal blow or aided the one who did, then, subject to proof of any other elements, that offender may be convicted. Where there are different hypotheses, but each consistent with guilt, a jury generally not need be unanimous as to which hypothesis is accepted. That is because the critical issue is satisfaction beyond reasonable doubt of the legal element(s).”

Submissions on the application

78. Having set out the relevant authorities relied upon by the parties in detail, the respective submissions can be succinctly summarised.

Mr Tadrosse’s submissions

79. The submission on behalf of Mr Tadrosse was a simple one. The Crown Case Statement indicates that the Crown only proposes to rely on EJCE which is a form of derivative liability. It was submitted that the Crown cannot put its case as one of EJCE as a matter of law on the facts of this case.

80. It was submitted that once it is accepted, as the Crown does, that the Crown cannot prove which of Mr Tadrosse or Mr Stephen is the principal offender who physically assaulted Mr and Mrs Saliba it is not open to the Crown to bring its case in reliance on principles of derivative liability because such liability is dependent on the guilt of a principal.

81. It was submitted that the Crown cannot prove beyond reasonable doubt that Mr Tadrosse is the principal or that Mr Stephen is the principal. This means that their liability as secondary parties cannot be proven beyond reasonable doubt either. As submitted by Mr Tadrosse, “the prosecution’s inability to prove who did attack Mr and Mrs Saliba also means it cannot prove who did not”.

82. It was submitted that if this argument was accepted, that does not mean that the Crown cannot bring its case. Rather, it was submitted that the Crown could rely on category (b) on the basis that murder and wounding were incidental crimes within the scope of Mr Tadrosse and Mr Stephen’s agreement. That basis of liability is principal rather than derivative and therefore there is no need to identify the principal offender on that basis of liability.

83. It was accepted that the Crown could bring its case under (c) without knowing the identity of the principal so long as the Crown could prove beyond reasonable doubt that the other persons played the role of a secondary offender.

84. In addition to [61] of Mitchell, passages from the judgment in Batak at [70], [93]-[94], [96], [189], [192] were relied upon in which comparisons were made between principles of accessorial liability and EJCE.

85. Although it was accepted that the Crown was permitted to put its case in alternative ways in the cases relied upon by the Crown extracted above at [69]-[77], it was submitted that the point now being made was not raised in those decisions.

Mr Stephen’s submissions

86. Mr Stephen adopted the submissions of senior counsel for Mr Tadrosse on this issue. It was submitted that given that the doctrine of EJCE is a form of derivative liability, the jury could not find either accused guilty as a secondary participant unless or until they were satisfied that a primary participant had committed the crimes of murder and wounding. As the Crown has not nominated a primary participant, and could not do so on the available evidence, it was submitted that the Crown should not be permitted to allege the liability of either accused according to principles of EJCE.

87. Further, it was submitted that the matter was further complicated by the possibility that there was a third offender. He suggested there were six different possibilities rather than simply two and went on to give examples as follows:

1. Mr Stephen is in the house and Mr Tadrosse is in the car;

2. Mr Tadrosse is in the house and Mr Stephen is in the car;

3. Both of them are in the car and there are other people in the house;

4. Mr Stephen is in the house, Mr Tadrosse is upstairs and there is another person downstairs committing the assaults.

Crown submissions

88. The Crown accepted that the doctrine of EJCE is derivative in nature but submitted this does not mean that the Crown is required to establish which of the accused physically did the acts which killed Mr Saliba and wounded Mrs Saliba.

89. Reliance was placed on the decisions summarised above at [69]-[77]. It was submitted that these cases show that even in a case of accessorial liability in which the Crown must establish the guilt of P1 beyond reasonable doubt, the Crown is permitted to run alternate cases as to which of the offenders is the principal.

90. It was submitted that the Crown should be permitted to base its case on proof of two possibilities being that both accused were either liable as the primary party or liable as a secondary party through the doctrine of EJCE, “with satisfaction beyond reasonable doubt that there was no reasonable hypothesis of any other possibility”: Baker v Smith at [19].

91. At [15] of its written submissions, the Crown proposed a hypothetical to demonstrate its position:

“An accused ‘T’ admits, during a controlled operation, that he had driven to a victim's home with an accomplice ‘X’ (a man he would not name) in order that they be robbed. T parked and X went to the house in order to carry out the robbery. T knew X was armed with a weapon and knew X would easily resort to extreme violence to achieve his desired purpose. That the Crown cannot nominate who was X, the primary party, should not impede T's liability in this hypothetical. The moral culpability of T, the secondary party, should be same, regardless of whether the Crown could nominate who was the primary party. Indeed, that EJCE could not be relied upon against T simply because the Crown could not nominate who was the primary party would be an absurdity.”

Consideration

92. The determination of this application does not turn on whether EJCE is now considered to be a doctrine of derivative liability; the Crown accepted it was. Rather, this application turns on what that means in a trial such as this and, in particular, to what extent principles of derivative liability addressing accessorial liability now apply to the doctrine of EJCE following the decision in Mitchell.

93. The starting point is to observe that although the judgment of Gordon, Edelman and Steward JJ in Mitchell at [61] described the doctrine of EJCE as “derivative” for the first time, it did not go on to explain what flowed from that new categorisation. It remains unclear to what extent aspects of the law governing accessorial liability now apply to principles of EJCE. Nor was the question of the extent to which the elements to be proved beyond reasonable doubt in an EJCE have changed given this new characterisation addressed.

94. A number of decisions of the Court of Criminal Appeal have adopted the statement of principle - in Mitchell at [61]. It would appear that in all of them the Crown accepted the categorisation of EJCE as now involving derivative rather than principal liability.

95. Although for my part, it remains unclear as to whether a majority of the High Court has in fact held that the doctrine of EJCE is now characterised as one of derivative rather than principal liability, I will proceed on that basis given the attitude of the parties in this application and the decisions of the Court of Criminal Appeal following Mitchell to which I have referred. To express doubt in that regard is not to say that I cannot see a basis to describe EJCE as some form of derivative liability; it certainly is appropriately described as a “parasitic” form of liability.

96. An offender’s liability in JCE category (a) is based on an agreement in which the acts of the co-offenders are attributed to him as part of what was agreed upon. Similarly, in relation to category (b) the incidental crime committed is still one that was within the scope of what was agreed. Category (c) (EJCE), on the other hand, differs in that it relies for proof of guilt on foresight of the incidental crime rather than agreement because the relevant acts were not part of what was agreed between the parties.

97. But simply accepting that EJCE is a form of derivative liability does not answer the question for determination.

To what extent do accessorial liability principles now apply to EJCE?

98. Although none of the Court of Criminal Appeal decisions which have followed Mitchell at [61] have addressed the issue that arises on this application, both the Crown and the accused relied on passages from the decision of the Court of Criminal Appeal in Batak in support of their respective positions. That decision concerned the interplay between principles of accessorial liability and constructive murder, but their Honours considered the overlap between principles of accessorial liability and JCE/EJCE as follows (at [92]-[93]):

“[92] The doctrine of extended joint criminal enterprise covers the situation where there is agreement to carry out a crime but that agreement does not encompass carrying out some further incidental crime. The accused may be liable for that incidental crime if they foresee that commission of that crime is a possibility and, with that foresight, continue to participate in the enterprise. For example, a participant in armed robbery may be liable for injury being inflicted on a victim by another participant if they foresaw that a weapon might be used even if they had all expressly agreed that it would not be: eg McAuliffe at 115. In Mitchell, Gordon, Edelman and Steward JJ stated that the nature of liability for EJCE differs from that for JCE. It does not involve agency or attribution of acts because the relevant acts were not part of what was agreed between the parties; liability is secondary and derivative (at [61]). The ‘liability of the accused is derived from, and dependent upon, the criminal liability of another for the foreseen crime that was not part of the agreement’ (ibid).

[93] In that respect, EJCE is similar to accessorial liability for accessories before or at the fact. A person who counsels, procures, aids or abets another commit a crime can also be liable for that crime. This type of liability differs from JCE in that it is not necessary for the Crown to establish any express or implied agreement to carry out the crime. As noted above (at [70]), however, where accessorial liability for accessories before the fact depends on establishing that the accused knew of the principal’s intention to carry out the crime and with that knowledge encouraged or assisted them – such that they are linked in purpose – in most cases it would also be possible to infer that there was agreement between them that the crime be carried out. And an agreement manifesting a JCE may itself be inferred from the parties’ conduct without the need for any express words: Miller at [4]. Nevertheless, the High Court has made clear that the doctrines of JCE and accessorial liability are distinct. A party may be liable under more than one doctrine. Thus the presence of agreement does not preclude the prosecution asserting that the person is liable as an accessory. (Emphasis added.)

99. Although it is to be accepted that EJCE is “similar” to accessorial liability, that is not to say that the elements are the same. As the Court in Batak went on to observe at [96]:

“Since McAuliffe the doctrines of JCE and EJCE have tended to become more commonly chosen prosecution pathways to guilt in cases involving criminal complicity to commit crimes than accessorial liability. EJCE may be available as a pathway to liability where accessorial liability is not available (and that is one of the reasons it has regularly been attacked in the High Court). If an accused merely foresaw that commission of a particular crime was a possible outcome of a particular course proposed by another party to the JCE then the accused could be liable under EJCE, but could not be liable as an accessory because they would not have the required actual knowledge of the intent to carry out the crime.”

100. As this passage confirms, the fact that both accessorial liability and EJCE are derivative in nature does not mean that they have the same elements. What does seem to be accepted is that they share the common element that each participant’s guilt is derived from the guilt of the principal.

101. Senior counsel for Mr Tadrosse accepted that it was not necessary to know the identity of P1 in a case of derivative liability, so long as the identity of P2 and their respective roles is clear. So much is consistent with what this Court (Button J with whom McNaughton J and Hulme AJ agreed) held in Jaghbir v R [2023] NSWCCA 175 at [195]-[197]:

“[195] But quite apart from that perhaps overly refined analysis, as I have shown, in my opinion there was no sense in which it was incumbent upon the Crown to prove beyond reasonable doubt that Mr Sleiman was, in fact, one of the principals in the first degree. As I have shown, the indictment did not aver it (and it was not the subject of demurrer or other criticism). Nor does the criminal law insist that, before an accessory before the fact can be found guilty, the principal in the first degree must be convicted, or even specifically identified: see s 346 of the Crimes Act 1900 (NSW); King v The Queen (1986) 161 CLR 423; [1986] HCA 59 Dawson J at 434 (Gibbs CJ, Wilson and Brennan JJ agreeing; Murphy, Mason and Deane JJ dissenting on other grounds).

[196] It is certainly uncontroversial, as was said on behalf of the applicant, that an accessory before the fact must know all of the physical and mental elements of the offence that is to be committed at the time of the intentional act of assistance done by the accessory. But that does not mean that the criminal law requires proof of knowledge by the accessory of the personal identity of an individual who is to be a perpetrator.

[197] And there are sound policy reasons for the law being that way: were it otherwise, a person could intentionally assist or encourage a very grave offence before its commission, well knowing that it was to occur, but so long as the accessory before the fact did not know the precise identity of the particular person who was to be the perpetrator, then the accessory would be exculpated.”

102. It seems to me that the principle that in a case of accessorial liability the identity of P1 need not be known, is relevant to the question for determination. The requirement in a case of accessorial liability that the Crown must prove beyond reasonable doubt that P1 is guilty of the principal offence, is a requirement that the principal offence was committed in the first place. If the principal offence was not committed, then P2 cannot be guilty.

103. As Button J explained in Fan, in a case of derivative liability the jury needs to be satisfied beyond reasonable doubt of the guilt of P1, before they can return a verdict against P2. That means that a P2 “get[s] the benefit” of any exculpatory matter pertaining to the shooter such as being found not criminally responsible due to a mental health impairment or cognitive impairment, or being under 10 years of age. In other words, the requirement that P1 is guilty before P2 can be guilty is a requirement that P1 be guilty, whoever he or she is.

104. In the present case, there is nothing to suggest that whoever P1 is in this trial, based on the anticipated evidence of Mrs Saliba, he is not guilty of intentionally inflicting (at least) grievous bodily harm on Mr Saliba and wounding Mrs Saliba. On the Crown case, it was either of the two accused.

105. Further and in any event, the issue raised on this application was held to be permissible and open to the Crown in Mohan v The Queen, R v Lowery & King (No 2) and Phan. Those cases all concerned principles of accessorial liability and were factually similar. In Phan, the proposed direction in a case when the Crown cannot prove which of two offenders was P1 and which was P2 was set out by Smart J at [111]-[112] as follows:

“[111] Upon the assumption that the Crown presents a case that the accused was either a principal offender or an aider and abettor, but it cannot say which, the jury should be instructed that before the accused can be convicted of murder the Crown has to prove the following beyond reasonable doubt:

(a) the deceased died as a result of a gunshot fired by the accused or [the co-offender]

(b) with intent to kill the deceased or inflict upon him grievous bodily harm, that is really serious bodily harm

(c) the accused and [the co-offender] were present at the time the crime was committed

(d) the accused and [the co-offender] knew all the essential facts or circumstances necessary to show that the crime was committed by the other.

(e) with that knowledge the accused or [the co-offender] intentionally assisted the other to commit the crime.

…

[112] It would also have to be explained that if the Crown cannot prove who fired the fatal shot or shots but is able to prove that the accused must have either fired the shot or aided and abetted, such accused is guilty of murder.” (Emphasis added.)

106. The facts in Phan were that there were three people in a car and one of them was shot. The Crown could not prove who was the shooter. The Court held that the re-trial could be run as one of accessorial liability on the basis that the accused was either P1 or P2; either he fired the shots or alternatively that he was an aider and abettor to the other person and liable as a principal in the second degree.

107. Although the elements of accessorial liability differ to those of EJCE, I can see no reason why, if in a case of accessorial liability (which is derivative and has as an element that P1 committed the offence) it is open to the Crown to bring the case in the alternative as to who is P1 or P2, the same principle would not apply to an EJCE as now characterised.

108. I am satisfied that, consistent with the authorities I have considered above in some detail, the elements of EJCE remain the same elements as prior to the decision in Mitchell at [61] with the addition that, given that the liability is derivative, the jury must be satisfied beyond reasonable doubt that P1 committed the offence, whether P1 be Mr Tadrosse (with Mr Stephen being P2) or whether P1 is Mr Stephen (with Mr Tadrosse being P2).

109. Based on principles of accessorial liability, it seems to me that, as a matter of policy, it cannot be the case that in a case based on EJCE if the Crown can prove that P1 was one or the other of a specified group but cannot say which one the Crown cannot bring its case on that basis. Whether the Crown can, on the evidence, in fact prove the elements of an EJCE case (including the scope of the agreement) whether the incidental offence was foreseen and that the P2 went along with the agreement, nonetheless, is a matter of evidence rather than law.

110. I have not overlooked the fact that there has been much criticism of EJCE as a form of complicity, particularly in its application to the crime of murder, given that the mental element to be foreseen is not necessarily that P1 will kill but rather that he or she will intentionally inflict grievous bodily harm. But the principle remains part of the common law of Australia and it remains open to the Crown to bring its case on this basis.

111. Finally, although any reference to the Criminal Trials Bench Book, published online by the Judicial Commission (“the Bench Book”), should be exercised with caution, there is nothing in the Bench Book to support the contention that the Crown ought not be permitted to run its case in the alternative because the offence is one of derivative liability. By way of observation, I note that the Bench Book in fact categorises (b) and (c) as both being forms of EJCE. An explanation is given for that at [2-740] as follows:

“For the purposes of the following suggested directions on extended criminal liability, (b) and (c) above are merged because the distinction may be confusing to a jury. Whether the crime committed is foreseen as a possible incident in carrying out the joint criminal enterprise, (b) above, or foreseen as a possible consequence of the commission of the joint criminal enterprise, (c) above, is not so significant a distinction as to require separate directions to meet those particular factual situations (except, potentially, in the case of alleged constructive murder: see [2-770]). The accused is criminally liable for the commission of the further offence, if they foresee the possibility of it being committed during the course of carrying out the joint criminal exercise no matter what the reason is for that foresight. The suggested directions use the term “additional crime” rather than ‘incidental crime’ or “consequential crime” to avoid the distinction which seems to be of theoretical more than of practical significance. It may be that, where the additional offence is viewed as incidental to the commission of the joint criminal enterprise, it will be more easily proved that the commission of that offence was foreseen as a possibility by a particular participant. The suggested directions are based on a scenario where the crime, the subject of the joint enterprise is committed and an additional crime is also committed.”

112. Although the Bench Book does not include as element of an EJCE that the jury must be satisfied beyond reasonable doubt that the principal offender is guilty before any of the co-accused can be found guilty, I am satisfied that is the case.

Elements to be proved

113. There was no discussion during the pre-trial argument as to what the elements of EJCE for murder based on principles of derivative liability are. It will be necessary for there to be agreement about that before the Crown opens its case on Monday. I would tentatively suggest that the elements the Crown must establish beyond reasonable doubt for murder are:

1. That Mr Tadrosse and Mr Stephen entered an agreement to assault the occupants of the house (that is, to cause them to apprehend immediate and unlawful violence) with intent to rob them of money or other valuable property, while being armed with an offensive weapon or instrument;

2. That each of Mr Tadrosse and Mr Stephen participated in that agreed crime;

3. That either Mr Tadrosse or Mr Stephen intentionally inflicted grievous bodily harm or intentionally killed Mr Saliba (the incidental crime);

4. That although the commission of the incidental crime by Mr Stephen or Mr Tadrosse was not within the scope of the agreement, it was contemplated by the other of them as a reasonable possibility that it might be committed by the other of them and that offender continued to participate in the agreement; and

5. If the jury is not satisfied beyond reasonable doubt of (3) then both accused must be acquitted. That is because unless the jury is satisfied beyond reasonable doubt that Mr Saliba was killed by one of them, the other cannot be guilty on the basis that he contemplated as a reasonable possibility that it might be committed and continued to participate in the agreement.

114. I would be assisted by submissions on this issue prior to the trial commencing.

Was there a third party to the agreement?

115. Finally, I have considered whether the prospect that there may have been a third party to the agreement changes my conclusion in this regard. The Crown Case Statement does not currently include any reference to there being a third person in the agreement in its description of how the Crown brings its case; the agreement is stated to be between the two co-accused. But, as outlined above, during the pre-trial submissions the Crown averted to the fact that there could have been a third party to the agreement and the Crown Case Statement at [6] may need to be amended to reflect that.

116. It is ultimately a matter for the Crown how it runs its case, but it seems to me that the prospect of a third person also being a party to the agreement is a question of evidence (whether the Crown can prove its case) rather than a matter of law arguing against the conclusion I have reached. If the Crown cannot exclude that there was a third person to the agreement it may well make the directions longer and also make the Crown case factually less straightforward to present to the jury. Despite this, for the reasons I have explained, it does not change my conclusion as to the relevant legal principles to be applied.

117. In the event that the Crown does amend its case to rely on an agreement that included a potential third participant accused, then clearly the elements of the offence at [113] above would need to be changed to reflect that.

118. Accordingly, I am satisfied that it is open to the Crown to run its case on extended joint criminal enterprise principles in this matter.

*********

Endnotes

Decision last updated: 20 August 2025