Dair v The State of Western Australia - [2008] WASCA 72

JURISDICTION : SUPREME COURT OF WESTERN AUSTRALIA

TITLE OF COURT : THE COURT OF APPEAL (WA)

CITATION: DAIR -v- THE STATE OF WESTERN AUSTRALIA [2008] WASCA 72

CORAM: STEYTLER P
MILLER JA
EM HEENAN AJA

HEARD: 5 FEBRUARY 2008

DELIVERED : 27 MARCH 2008

FILE NO/S: CACR 33 of 2007
CACR 32 of 2007

BETWEEN: DEVRON DAIR
Appellant
AND
THE STATE OF WESTERN AUSTRALIA
Respondent

ON APPEAL FROM:
Jurisdiction              :  DISTRICT COURT OF WESTERN AUSTRALIA
Coram  :HEALY DCJ
File No  :IND 576 of 2004

Catchwords:

Criminal law - Appeal against conviction - Unlawfully doing grievous bodily harm with intent to maim, disfigure, disable or do grievous bodily harm - Admissibility of evidence relating to identification - Circumstantial evidence and not positive identification - Admissibility of evidence of past convictions - Section 31A Evidence Act1906 (WA) - Evidence of recognition

Evidence - Admissibility of evidence falling short of identification - Circumstantial evidence - Appropriate directions - Admissibility of past convictions - Section 31A Evidence Act1906 (WA) - Meaning of 'significant probative value' - Whether probative value outweighed risk of an unfair trial - Evidence of recognition - Extent to which witness could give evidence of special features relating to appellant

Criminal law - Sentencing - Unlawfully doing grievous bodily harm with intent to maim, disfigure, disable or do grievous bodily harm - Sentence of 8 years' imprisonment without parole - Whether manifestly excessive - Whether sentencing judge erred in failing to take into account time previously spent in custody - Appellant in custody 3 years 4 months prior to arrest and taken into custody for present offence

Legislation:

Criminal Code (WA)
Criminal Procedure Act 2004 (WA), s 98(2)(a)
Evidence Act 1906 (WA), s 31A
Sentencing Act (WA)
Sentencing Legislation Amendment and Repeal Act 2003 (WA)

Result:

Appeal against conviction allowed

Category: A

Representation:

Counsel:

Appellant: Mr M R Gunning
Respondent: Mr S E Stone

Solicitors:

Appellant: Gunning Young
Respondent: Director of Public Prosecutions (WA)

Case(s) referred to in judgment(s):

Australian National Industries Ltd v Spedley Securities Ltd (in liq) (1992) 26 NSWLR 411
Azarian v The State of Western Australia [2007] WASCA 249
BRS v The Queen [1997] HCA 47; (1997) 191 CLR 275
Davies v The King (1937) 57 CLR 170
Di Lena v The State of Western Australia [2006] WASCA 162; (2006) 165 A Crim R 482
Director of Public Prosecutions v P [1991] 2 AC 447
Domican v The Queen (1992) 173 CLR 555
Donaldson v The State of Western Australia [2005] WASCA 196; (2005) 31 WAR 122
DPP v Boardman [1975] AC 421
Festa v The Queen [2001] HCA 72; (2001) 208 CLR 593
Fordham (1997) 98 A Crim R 359
Gipp v The Queen [1998] HCA 21; (1998) 194 CLR 106
Goldsmith v Sandilands [2002] HCA 31; (2002) 76 ALJR 1024
Harriman v The Queen (1989) 169 CLR 590
Hoch v The Queen (1988) 165 CLR 292
Jacara Pty Ltd v Perpetual Trustees WA Ltd [2000] FCA 1886; (2000) 106 FCR 51
Jarvis v The Queen (1993) 20 WAR 201
Johnson v Johnson [2000] HCA 48; (2000) 201 CLR 488
Kilby v The Queen [1973] HCA 30; (1973) 129 CLR 460
KRM v The Queen [2001] HCA 11; (2001) 206 CLR 221
Libke v The Queen [2007] HCA 30; (2007) 235 ALR 517
Lock (1997) 91 A Crim R 356
Lockyer (1996) 89 A Crim R 457
Mill v The Queen (1988) 166 CLR 59
Murphy v The Queen (1994) 62 SASR 121
Neville v The Queen [2004] WASCA 62; (2004) 145 A Crim R 108
Noto v The State of Western Australia [2006] WASCA 278; (2006) 168 A Crim R 457
Palmer v The Queen [1998] HCA 2; (1998) 193 CLR 1
Pfennig v The Queen [1995] HCA 7; (1995) 182 CLR 461
Phillips v The Queen [2006] HCA 4; (2006) 225 CLR 303
Pitkin v The Queen [1995] HCA 30; (1995) 69 ALJR 612
Postiglione v The Queen [1997] HCA 26; (1997) 189 CLR 295
R v AH (1997) 42 NSWLR 702
R v Alexander [2002] VSCA 183; (2002) 6 VR 53
R v Best [1998] 4 VR 603
R v Delgado‑Guerra [2001] QCA 266; [2002] 2 Qd R 384
R v Mateiasevici [1999] VSCA 120; [1999] 3 VR 185
R v Straffen [1952] 2 QB 911
R v Tektonopoulos [1999] VSCA 93; [1999] 2 VR 412
Raybos Australia Pty Ltd v Tectran Corporation Pty Ltd [No 9] (Unreported, NSWCA, 27 November 1990)
Reid v The Queen [1990] 1 AC 363
Smith v The Queen [2001] HCA 50; (2001) 206 CLR 650
Sutton v The Queen (1984) 152 CLR 528
The State of Western Australia v Osborne [2007] WASCA 183
Thompson v The Queen (1989) 169 CLR 1
WRC [2002] NSWCCA 210; (2002) 130 A Crim R 89
Zaknic Pty Ltd v Svelte Corporation Pty Ltd (1995) 61 FCR 171

STEYTLER P: I have had the advantage of reading the judgment of Miller JA. I will adopt his comprehensive statement of the facts giving rise to this appeal against conviction and sentence.
The appellant was convicted, after a trial, on a charge of unlawfully doing grievous bodily harm with intent to maim, disfigure, disable or do grievous bodily harm. The victim was a police officer, Constable Normie Philip Giocas. The appellant was sentenced to a term of 8 years' imprisonment.

Circumstances of the offence

On the afternoon of 3 April 2003, Mr Thiam Hoi Yap walked to his car, which was parked in the City Library carpark in Northbridge. When he got there, he found the driver's door open and a man sitting in the driver's seat. The man grabbed a bag from the passenger side of the car. He jumped out of the car and ran to an exit stairwell. Mr Yap ran after him but, when threatened, gave up his pursuit.
Constable Giocas was in the carpark at the time. He was off duty. After speaking briefly to Mr Yap, he chased the offender. He called upon the offender to stop, and identified himself as a police officer. The offender continued to run. Constable Giocas chased him into the plaza of the Perth Cultural Centre. The offender turned and faced Constable Giocas. He had a knife. He took some steps towards Constable Giocas and then stabbed him in the neck. The offender escaped.
Constable Giocas was seriously wounded. His superior thyroid artery had been partially severed. Were it not for emergency surgery, he might have lost his life. He made a full recovery.
The scene of the crime was searched. Police found a screwdriver in the stairwell of the carpark through which the offender had run. Marks on Mr Yap's car were consistent with the use of that screwdriver.

Witnesses who saw the offender

Apart from Mr Yap and Constable Giocas, there were four witnesses who saw the offender. However, none of the witnesses (including Mr Yap and Constable Giocas) could make a positive identification. Each gave a description of the offender. As is apparent from Miller JA's reasons, these descriptions varied in a number of respects, some of them substantial.
One of the witnesses, Deborah Symington, was taken to Hakea Prison on 22 May 2003. The appellant was then being held there. She sat in a room and watched a number of men walk past. She was asked if she could identify the offender. The appellant was one of the men who walked past the room. She picked him out of the group. However, she said that she could only give him five and a half or six out of 10 as being the person that she had seen in the car park.
Another witness, Melody Dawes (formerly Melody Woollard), was taken to police headquarters on 22 April 2003. There she assisted a police officer to make up an identikit picture of the offender. She was also shown a photo board containing a number of photographs. At first, she could not identify anybody from the board. Then, when asked whether there was any person on the board who looked like the person she had seen, she picked out two photographs. One of these was a photograph of the appellant. She said that the photograph of the appellant stood out 'more' than that of the other person that she had picked out. She said that the appellant looked 'similar' but that she could not be sure that he was the offender.
A third witness, Valma Hughes, described the offender as having 'a strange running gait'. She said that, as he ran, 'his left leg came in … quite a lot' towards his right knee.
Ms Maria Carvalho, the fourth witness, saw the offender through the rear vision mirror of her car as she was driving through the car park. However, the description that she gave of the offender was so general as to be of little assistance.
Constable Giocas assisted a sketch artist to draw a sketch of the person who had stabbed him. He also assisted in the preparation of a computer composite sketch of that person. On 9 May 2003 he went to Hakea Prison. A number of men, including the appellant, walked past the room in which he was sitting. He did not immediately identify anyone. However, he eventually picked out the appellant. In his evidence, he said:
His height was similar, his hair colour was similar, although his hair was shorter, and his facial features ‑ more likely his jaw line was similar but he was of a heavier build and … there was something about him and that's what I thought.
Mr Yap, too, attended an identification parade at Hakea Prison. He did so on 22 May 2003. He could not recognise the offender. This was so even though the appellant was the only Caucasian person dressed in prison greens who walked past him. Only two Caucasian people were in the parade.

Other evidence

It is obvious that the appellant could not have been convicted on the strength of this identification evidence alone. However, there was other evidence.

Ian Strachan

The most important prosecution witness was Ian Strachan. He said that he had met the appellant in February 2003. The two of them were amphetamine users. After meeting the appellant, he saw him 'pretty much every day'. He was asked, during cross‑examination, whether there had been times when he had assisted the appellant by putting him in touch with people who bought stolen goods. He acknowledged that he had done so. He was able to produce telephone numbers for the appellant that had been stored in his mobile telephone.
On 3 April 2003 Mr Strachan was watching television. He saw a news bulletin relating to an attack on a policeman. In the course of the bulletin, surveillance footage from the car park was shown. The bulletin also displayed a hand‑drawn identikit picture. Mr Strachan said that he had immediately identified the appellant from the surveillance footage. This was because he 'got a pretty good side profile'. It was also because the appellant had 'a very distinctive walk'. He said that the identikit sketch was 'pretty close to how [the appellant] looked ... at the time'. He said that the appellant's hair was 'probably a little bit shorter on top'.
He met the appellant later that evening. He noticed that the appellant had shaved his head. He asked the appellant about the stabbing. He said that the appellant acknowledged that he had stabbed the police officer. The appellant produced the knife that he had used. It had what Mr Strachan believed was dried blood still on it. Mr Strachan said that the appellant told him that he had been caught breaking into a vehicle and that 'they tried to stop him'. He said that that was why he 'stabbed the copper'. Mr Strachan said that he told the appellant that the police officer could have died. The appellant responded by saying, 'Yeah, well I hope the fucker does'. The appellant told Mr Strachan that he had shaved his head so as 'to change his appearance to hide'.
Mr Strachan said that the appellant had shown him how he had stabbed Constable Giocas. He said that this had been done with a 'backhand' motion. This differed from the description given by Constable Giocas, who described a round‑arm blow. It also differed from the description given by Ms Dawes, who saw a bent elbow, a little above shoulder height, and the assailant's fist moving forward.
Mr Strachan said that the appellant always carried a backpack full of tools. He saw the appellant repacking the backpack on one occasion. It contained small crowbars, a big bunch of keys, screwdrivers, pliers and other 'bits and pieces'. It also contained a number of 'little immobiliser tags'.
Mr Strachan first told police what he knew about the stabbing of Constable Giocas on 28 April 2003, some 3 1/2 weeks after the alleged confession was made to him by the appellant. Mr Strachan was then in Esperance. He had been there for rehabilitation in respect of his drug addiction. He was interviewed by the police about the stabbing of a woman who was a friend of his. The police were looking for information about that stabbing. During the interview, Mr Strachan volunteered that he knew who had stabbed Constable Giocas. He made a formal statement on the following day. He had said nothing to the police earlier because he was afraid to do so. However, he felt safe in Esperance.
Mr Strachan acknowledged that, when he made his statement to police, he was subject to a suspended sentence of imprisonment and a community service order. These had been imposed in respect of a conviction for fraud relating to cars that he had hired, using false identification.

Evidence of convictions for similar offences

Evidence was also made available to the jury which established that the appellant had been convicted of four other offences. This evidence had been ruled admissible by Fenbury DCJ after a preliminary hearing which had concluded on 9 January 2006, more than 12 months prior to the trial. Three of the convictions proved by this evidence preceded the stabbing of Constable Giocas. The fourth was subsequent to it.
In February 1998 the appellant stole a car. He was pursued by police, while driving the stolen car. After ramming the police car with the stolen car a number of times, he abandoned the car. He was chased on foot. When cornered, he sprayed three of the pursuing officers with pepper spray. He bit the finger of one of them, causing him to have a number of stitches.
In September 1998 he stole a car, using a screwdriver. Once again, he became involved in a high speed car chase. He abandoned the stolen car. When he was eventually caught, he resisted and struggled violently. Screwdrivers, ignition bolts and car keys were found in his possession.
On 9 February 2002 he stole a car. Again, he became involved in a pursuit. He rammed the pursuing police car. He also stopped the stolen car suddenly, forcing the police car to collide with it. When the police approached the stolen car, he locked its doors. The police forcibly entered the stolen car. He punched and kicked police officers when they tried to arrest him.
The offence committed by the appellant subsequent to the stabbing of Constable Giocas took place on 16 April 2003. This, too, involved the theft of a car and a high speed police chase. The appellant stopped the stolen car during the chase and reversed it into the pursuing police car. The police car was so badly damaged that it could no longer be driven. A second police car took up the chase. Again, the appellant reversed into it. He eventually abandoned the stolen car and ran off on foot. He was chased by a bystander. The appellant sprayed the bystander with capsicum spray. He brandished a knife and capsicum spray at two other bystanders. He was eventually cornered and police apprehended him. The knife that he had brandished was found in a grassed area where he had attempted to conceal it. A subsequent search of premises in which the appellant had stayed revealed a number of backpacks. Some of these contained big bunches of keys and a number of tools, including screwdrivers. Knives were also seized. At the ensuing sentencing hearing, the appellant's counsel told the court that the appellant 'always carries a pocket knife'.
The items seized by police after the offence committed on 16 April 2003 were subjected to DNA testing. No blood belonging to Constable Giocas was found on any of them. No fingerprints matching those of the appellant had been found on Mr Yap's car.
In deciding to admit the evidence of the four convictions pursuant to s 31A of the Evidence Act 1906 (WA), Fenbury DCJ found that it had significant probative value. He said, in this respect (ts 109):
As counsel put it, the strength of the probative force of evidence of this kind lies in the fact that it reveals underlying unity, system or pattern, such that it raises … as a matter of common sense and experience the objective improbability of some event having occurred other than as alleged by the prosecution.
Fenbury DCJ also seemingly approved the following submissions that had been made on behalf of the prosecution (ts 109, 110):
It was submitted that Strachan's evidence of the accused [sic] from the carpark surveillance footage must be challenged on the basis that it's either untruthful or mistaken, and it's submitted that the propensity evidence demonstrates that it's objectively improbable; that Strachan was simply concocting the making of the admissions; that if made, the admissions were false ‑ that is, if made by the accused the admissions were false ‑ and that Strachan's identification of the accused from the surveillance footage is mistaken or dishonestly made.
It was put at paragraph 44 the propensity evidence is admissible to support the reliability of the witness' identification evidence, … 'in that it is objectively improbable given the underlying unity, system or pattern demonstrated by it that each of the witnesses would be mistaken in selecting the accused as the person best resembling the offender.' It was put that the propensity evidence is also admissible to support the meaning the State attributed to the statements made by the accused in the Arunta phone calls [a reference to tape‑recorded telephone calls made by the appellant from prison] …
It was suggested that it was objectively improbable that a person accused of the present offence with the background of this accused would make those statements if he were not the offender. Then it was finally put that the risk of an unfair trial can be adequately dealt with by judicial direction which would need to be of course a firm direction, which goes without saying.

Tape‑recorded telephone conversations

The State led evidence of two tape‑recorded telephone conversations that had been made by the appellant from Albany Regional prison, after he had been charged with the present offence.
The first of these was a telephone call made to Mr John Chalupa. In the course of it, the following exchanges took place:
APPELLANT:   Yeah, they're charging me with that stabbing.
CHALUPA:Yeah, I see that in the paper, yeah.
APPELLANT:   Yeah. Yeah, I don’t know, there's some shonky shit going on.
…
APPELLANT:   So, one of the coppers who initially first come to interview me, he already knows what I look like, right, and he's, um, brought a witness, namely that one that was stabbed, into the prison, right? And ‑ ‑ and pointed me out as I've gone and seen my lawyers, on the side, you know what I mean?
CHALUPA:Oh, yeah.
APPELLANT:   Oh, there's all this other shit going on, too, you know? So ‑ ‑
CHALUPA:Mm.
APPELLANT:   ‑ ‑ I'm just going to see what happens and that, you know, and ‑ ‑
CHALUPA:Yeah.
APPELLANT:   ‑ ‑ hope ‑ ‑ hopefully they've slipped up, you know?
CHALUPA:Mm.
APPELLANT:   With the shit they've done.
…
APPELLANT:   So, I don't know what I'm looking at. I've got to ‑ ‑ um, as it is, I've got 4 and a half with parole.
CHALUPA:Mm.
APPELLANT:   I've got another year and a half to do already, ah, without this other shit yet, which I've still got to, you know, sort out.
CHALUPA:Yeah.
APPELLANT:   Um, see what happens to that, you know. I might be looking at a while, who knows?
CHALUPA:Mm.
APPELLANT:   Yeah. Oh well, live and learn the hard way, eh?
The second telephone call was made by the appellant to his father, George Dair. In the course of it, the following exchanges took place:
APPELLANT:   Yeah, I got ‑ ‑ I still got a GBH charge to go up on.
GEORGE
DAIR:When's that?
APPELLANT:   Um, well February 19th is going to be a video link from down here.
GEORGE
DAIR:Yeah?
APPELLANT:   I'm waiting to see what evidence they got.
GEORGE
DAIR:Yeah.
APPELLANT:   Um, and then I may have to go up there again, probably about April maybe.
GEORGE
DAIR:Oh, yeah.
APPELLANT:   Just probably Casuarina for a week, to go to court, then come back.
GEORGE
DAIR:Mm. Oh, yeah.
APPELLANT:   So don't know what's happening there.
GEORGE
DAIR:Mm.
APPELLANT:   Might be looking at a bit longer.
GEORGE
DAIR:I'd say you will be.
APPELLANT:   Yeah.
GEORGE
DAIR:Hey?
APPELLANT:   Yeah.
GEORGE
DAIR:Yeah, well, you do silly things.
APPELLANT:   Yeah. Shit happens.
GEORGE
DAIR:Yeah. It happens all right, doesn't it?
APPELLANT:   Yeah. Learn the hard way.

At the trial, the prosecutor contended that these conversations amounted to admissions by the appellant.

Grounds 1, 2 and 3 of the appeal against conviction

Grounds 1, 2 and 3 of the grounds of appeal against conviction (dealt with more fully by Miller JA) essentially contend that the evidence of Constable Giocas, Ms Symington and Ms Dawes should not have been admitted. As Miller JA points out, these grounds were supported by two contentions. The first is that there was 'a real danger' that the jury would use the evidence 'as identification evidence' when it did no more than identify 'someone with similar features'. The second is that the identification procedures were deficient.
In the case of Constable Giocas, the procedure is said to have been deficient for three reasons. The first is that the identification took place in a prison (having what was described as a 'rogues' gallery effect'). The second is that others involved in the identification parade had beards or were of different nationalities to the appellant, making it more likely that he would be picked out. The third is that there was a danger that there had been 'contamination' of the identification by police officers who knew of the appellant's identity.
In the case of Ms Symington, there are said to have been four deficiencies. The first is that there was a 'reduced number of participants' in the parade and these had different 'traits' to the appellant, making it more likely that he would be identified. The second is that the identification parade was held in prison surroundings. The third is that there was a danger of police contamination, for reasons similar to those given in respect of Constable Giocas' tentative identification of the appellant. The fourth is that there was a danger of 'contamination' because there was a computer‑generated 'identikit' picture of the offender on a board in the room in which Ms Symington sat in order to make the identification.
In the case of Ms Dawes, the procedure was said to have been deficient because, after she had failed to make any identification from the photo board while being video‑taped, the video recorder was turned off. There was a short interval before the video‑taping resumed. Only then did Ms Dawes pick out two of the photographs. Counsel for the appellant contends that there is no way of knowing what took place in the intervening period. He also contends that Ms Dawes' evidence is further weakened by the fact that her description of the offender did not match the appearance of the appellant.
In his oral submissions, counsel for the appellant acknowledged that the evidence of each of these three witnesses, to the extent that it assisted in the identification of the appellant, was relevant. However, he contended that the evidence should have been excluded in the exercise of a discretion because its prejudicial value exceeded its probative value.
It is settled that evidence falling short of positive identification may nevertheless be of significance, having regard to the whole of the evidence: Festa v The Queen [2001] HCA 72; (2001) 208 CLR 593 [11] (Gleeson CJ). In the same case McHugh J [56] described this kind of evidence as 'circumstantial identification evidence'. He said that, because it is evidence that the general appearance or some characteristic or propensity of the accused is similar to that of the offender, it differs from positive identification evidence in that the witness does not claim to recognise the accused as the offender. However, he said that the evidence is nonetheless admissible as 'proof of a circumstance ‑ usually, but not always, weak ‑ that with other evidence may point to the accused as the person who committed the crime'. (See also Hayne J [217] ‑ [219], Callinan J [249] ‑ [253] and Kirby J (dissenting) [164], [165]; Smith v The Queen [2001] HCA 50; (2001) 206 CLR 650 [14] ‑ [15] (Gleeson CJ, Gaudron, Gummow & Hayne JJ).)
Of course, the discretion to exclude evidence applies as much to circumstantial identification evidence as it does to positive identification evidence: Festa [63] (McHugh J). However, the fact that circumstantial evidence would ordinarily have less probative value than positive identification evidence does not mean that it should be excluded. Its probative value may still exceed its prejudicial value. As McHugh J pointed out in Festa [65], because 'circumstantial identification evidence is usually no more presumptively prejudicial than other forms of circumstantial evidence, the occasions for its exclusion under the unfairness rule are likely to be fewer than the occasions for excluding positive identification evidence'. Where it is made plain to the jury that the evidence is not positive identification evidence, but merely evidence of similarity, and where defects in the evidence are pointed out to the jury, it is difficult to see why there should be any prejudicial effect sufficient to justify its exclusion. As Miller JA has pointed out, in this case the trial judge made it plain enough that the evidence was not evidence of positive identification. I agree with and adopt what Miller JA has said in that respect when dealing with grounds 1, 2, 3 and 5.
As to the complaint that those who organised the identification parades did so in such a way as to make the appellant stand out, any deficiencies (and I should say that the procedures adopted were a good deal less than desirable) would have been obvious to the jury. They saw the videotapes of the identification parades. They also had their attention directed to the deficiencies by the then counsel for the appellant. It is significant, in this respect, that there was no complaint at the trial that the trial judge (who also pointed out some of the deficiencies complained of) should have said anything more than he did concerning the procedures that had been adopted.
Also, there is no evidence of any 'contamination' by police officers, in the sense of identifying the appellant as the person who should be picked out. None of the witnesses suggested that anything of that kind had happened. There is no apparent reason to doubt their evidence in that respect, more especially in circumstances in which none of them made a positive identification. In the case of Ms Symington, her evidence was that she did not see the identikit picture of the offender on the board in the room from which she was required to view the parade.
Nor is there anything unduly untoward, in this case, in the fact that the identification parades (if that is a correct description of them) took place at a prison. That would have revealed only that the suspected offender was being held in custody. In these circumstances, and especially having regard to the fact that it was repeatedly made plain to the jury that none of the witnesses positively identified the appellant, I am not persuaded that the prejudicial value of the evidence exceeded its probative value. Grounds 1, 2 and 3 have not been made out.

Grounds 4 and 7 of the appeal against conviction

Ground 4 challenges the admission of evidence of the appellant's convictions in respect of other offences. Ground 7 challenges the trial judge's directions concerning this evidence.
The only legitimate purpose for which the evidence of other convictions could have been adduced in this case was one involving propensity reasoning. The evidence enabled the jury to infer that, because the appellant had a tendency to steal cars and then to use violence in the course of resisting arrest when apprehended, it was more probable that he had been correctly identified as the offender on this occasion than if he had not had this tendency.

Propensity evidence and the common law

The common law has always treated evidence of this kind with considerable caution. Under the common law, evidence of propensity or disposition is admissible only if it is sufficiently highly probative of a fact in issue as to outweigh the prejudice it might cause. It ought not be admitted at common law if there is a rational view of the evidence which, when the evidence of propensity or disposition is considered with other relevant evidence, is inconsistent with the guilt of the accused: Pfennig v The Queen [1995] HCA 7; (1995) 182 CLR 461, 481 ‑ 483 (Mason CJ, Deane & Dawson JJ); Hoch v The Queen (1988) 165 CLR 292, 296 (Mason CJ, Wilson & Gaudron JJ); BRS v The Queen [1997] HCA 47; (1997) 191 CLR 275, 298 ‑ 299 (Gaudron J); Phillips v The Queen [2006] HCA 4; (2006) 225 CLR 303 [9] (Gleeson CJ, Gummow, Kirby, Hayne & Heydon JJ)
Similar fact evidence is a form of propensity evidence. In Pfennig, Mason CJ, Deane and Dawson JJ said (464):
There is no one term which satisfactorily describes evidence which is received notwithstanding that it discloses the commission of offences other than those with which the accused is charged. It is always propensity evidence but it may be propensity evidence which falls within the category of similar fact evidence, relationship evidence or identity evidence. Those categories are not exhaustive and are not necessarily mutually exclusive. The term 'similar fact' evidence is often used in a general but inaccurate sense.
The probative force of similar fact evidence 'lies in the fact that it discloses some feature which raises, as a matter of common sense and experience, the objective improbability of its bearing an explanation consistent with the accused's innocence of the offence charged': BRS (298) (Gaudron J). That may be because it discloses strikingly similar facts, some unusual feature common to the events in question or some underlying unity, system or pattern: Hoch (294 ‑ 295) (Mason CJ, Wilson & Gaudron JJ); Pfennig (482) (Mason CJ, Deane & Dawson JJ) and BRS (299) (Gaudron J). However, these indicators are not exhaustive of the kinds of evidence that have special probative value: BRS (299) (Gaudron J); Pfennig (482, 484) (Mason CJ, Deane & Dawson JJ); Phillips [54].
Heydon's Cross on Evidence (7th Aust ed, 2004) suggests [21095] that the probative force of similar fact evidence depends upon three principal factors:
(1)the cogency of the evidence showing the accused's bad disposition;
(2)the extent to which such evidence supports the inference sought to be drawn from it; and
(3)the degree of relevance of that inference to some fact in issue in the proceedings.
He points out that these factors interact and that, to some extent, strength in one can compensate for weakness in another.
The real basis upon which the admission of similar fact evidence is permitted is that it must be strongly probative of a fact in issue: Heydon, [21120]. Heydon also points out (at [21120]) that it is the task of the judge not just to list similarities which appear striking, but to 'seek out and assess the logical nexus between the evidence in question and the facts in issue to determine whether as a matter of common sense and experience the evidence raises the objective improbability of some event occurring otherwise than as alleged by the Crown'. (See also A Ligertwood, Australian Evidence (4th ed 2004) [3.47].)
When propensity evidence is excluded under the common law, it is usually because the probative value of that evidence is outweighed by the danger that the jury will use it to support an impermissible line of reasoning, being one to the effect that, because the accused has a tendency to commit crimes, or crimes of a particular kind, it is probable that he committed the crime charged. However, the distinction between permissible and impermissible reasoning becomes very fine in a case in which similar fact evidence is led as evidence of identification of the accused. In such a case the evidence is led for the very purpose of inviting the jury to infer that, because the accused has a tendency to commit crimes of that character, it is more likely that he committed the present crime: see, for example, R v Straffen [1952] 2 QB 911 and Pfennig. Consequently, where the similar fact evidence is the only, or principal, evidence sought to be led in support of the guilt of the accused person, the evidence must be compelling before it will be admitted. It would otherwise be impossible to suggest that there was no rational view of the evidence that was consistent with the innocence of the accused.
In Director of Public Prosecutions v P [1991] 2 AC 447, the House of Lords drew a distinction between similar fact evidence adduced so as to identify an offender and similar fact evidence adduced for other reasons. Lord Mackay LC (with whom the other Law Lords were in agreement) said (462) that, ordinarily, similar fact evidence 'may take many forms and while these forms may include "striking similarity" in the manner in which the crime is committed, consisting of unusual characteristics in its execution, the necessary relationship is by no means confined to such circumstances'. He also said that 'Relationships in time and circumstances other than these may well be important relationships in this connection'. However, he went on to say (also at 462):
Where the identity of the perpetrator is in issue, and evidence of this kind is important in that connection, obviously something in the nature of what has been called in the course of the argument a signature or other special feature will be necessary.
This distinction has not been accepted in Australia. In R v Delgado‑Guerra [2001] QCA 266; [2002] 2 Qd R 384 [19], Thomas JA (with whom McMurdo P and Helman J agreed) said, after considering the English cases (including DPP v P):
It is difficult to think that some special requirement or different test should be applied in relation to the use of propensity evidence for proof of identity than for the other purposes for which it may be admitted … The essential question is whether the evidence is cogent enough to be admitted as proof of identification of the offender. The suggested 'signature' test in my view raises an unnecessary distinction and ought not to be followed.

Section 31A of the Evidence Act

Section 31A was introduced in the Evidence Act 1906 by s 13 of the Criminal Law Amendment (Sexual Assault and Other Matters) Act 2004 (WA). As Miller JA has said, the introduction of that provision was designed to give the courts greater capacity to admit propensity and relationship evidence: Western Australia, Parliamentary Debates, Legislative Assembly, 30 June 2004, 4608 (Mr JA McGinty, Attorney General). The section reads as follows:
31A.Propensity and relationship evidence
(1)In this section ‑
'propensity evidence' means ‑
(a)similar fact evidence or other evidence of the conduct of the accused person; or
(b)evidence of the character or reputation of the accused person or of a tendency that the accused person has or had;
'relationship evidence' means evidence of the attitude or conduct of the accused person towards another person, or a class of persons, over a period of time.
(2)Propensity evidence or relationship evidence is admissible in proceedings for an offence if the court considers ‑
(a)that the evidence would, either by itself or having regard to other evidence adduced or to be adduced, have significant probative value; and
(b)that the probative value of the evidence compared to the degree of risk of an unfair trial, is such that fair minded people would think that the public interest in adducing all relevant evidence of guilt must have priority over the risk of an unfair trial.
(3)In considering the probative value of evidence for the purposes of subsection (2) it is not open to the court to have regard to the possibility that the evidence may be the result of collusion, concoction or suggestion.
Miller JA has pointed out that the wording of s 31A(2)(b) has been taken from the dissenting judgment of McHugh J in Pfennig (529).

Other legislation

A different statutory approach has been taken in other States. So, for example, in Victoria s 398A(2) of the Crimes Act 1958 (Vic) provides that propensity evidence relevant to facts in issue in a proceeding for an offence 'is admissible if the court considers that in all the circumstances it is just to admit it despite any prejudicial effect it may have on the person charged with the offence'. That section adopts the approach taken by the English common law in cases such as DPP v P, rather than that adopted under Australian common law: R v Best [1998] 4 VR 603, 609; R v Tektonopoulos [1999] VSCA 93; [1999] 2 VR 412 [18]. The Victorian Court of Appeal has held that subs 398A(2) enacted a broad exclusionary rule extending to any evidence of discreditable conduct 'which may cause the jury to infer that a person who has been responsible for or involved in these acts is likely by reason of that fact to have committed the crime charged': Best, 608 (Callaway JA); Tektonopoulos [21]; R v Mateiasevici [1999] VSCA 120; [1999] 3 VR 185. The subsection applies whether or not the evidence is tendered for propensity purposes, although the purpose for which it is admitted remains relevant when considering whether it is 'just' to admit it: Tektonopoulos [24] ‑ [25] (Winneke P); R v Alexander [2002] VSCA 183; (2002) 6 VR 53 [43] (Winneke P).
On the other hand, the Evidence Acts of the Commonwealth, New South Wales and Tasmania (each of which is based upon the Australian Law Reform Commission Draft Bill) distinguish between 'tendency' evidence and 'coincidence' evidence. This is done in s 97, s 98 and s 101 of those 'Uniform Acts', which relevantly read as follows:
97.The tendency rule
(1)Evidence of the character, reputation or conduct of a person, or a tendency that a person has or had, is not admissible to prove that a person has or had a tendency (whether because of the person's character or otherwise) to act in a particular way, or to have a particular state of mind, if:
(a)the party adducing the evidence has not given reasonable notice in writing to each other party of the party's intention to adduce the evidence, or
(b)the court thinks that the evidence would not, either by itself or having regard to other evidence adduced or to be adduced by the party seeking to adduce the evidence, have significant probative value.
(2)Paragraph (1)(a) does not apply if:
(a)the evidence is adduced in accordance with any directions made by the court under section 100, or
(b)the evidence is adduced to explain or contradict tendency evidence adduced by another party.
98.The coincidence rule
(1)Evidence that 2 or more related events occurred is not admissible to prove that, because of the improbability of the events occurring coincidentally, a person did a particular act or had a particular state of mind if:
(a)the party adducing the evidence has not given reasonable notice in writing to each other party of the party's intention to adduce the evidence, or
(b)the court thinks that the evidence would not, either by itself or having regard to other evidence adduced or to be adduced by the party seeking to adduce the evidence, have significant probative value.
(2)For the purposes of subsection (1), 2 or more events are taken to be related events if and only if:
(a)they are substantially and relevantly similar, and
(b)the circumstances in which they occurred are substantially similar.
(3)Paragraph (1)(a) does not apply if:
(a)the evidence is adduced in accordance with any directions made by the court under section 100, or
(b)the evidence is adduced to explain or contradict coincidence evidence adduced by another party.
…
101.Further restrictions on tendency evidence and coincidence evidence adduced by prosecution
(1)This section only applies in a criminal proceeding and so applies in addition to sections 97 and 98.
(2)Tendency evidence about a defendant, or coincidence evidence about a defendant, that is adduced by the prosecution cannot be used against the defendant unless the probative value of the evidence substantially outweighs any prejudicial effect it may have on the defendant.
(3)This section does not apply to tendency evidence that the prosecution adduces to explain or contradict tendency evidence adduced by the defendant.
(4)This section does not apply to coincidence evidence that the prosecution adduces to explain or contradict coincidence evidence adduced by the defendant.

This legislation was considered by the Court of Criminal Appeal in New South Wales in WRC [2002] NSWCCA 210; (2002) 130 A Crim R 89. Hodgson JA (with whom Greg James & Kirby JJ were relevantly in agreement) expressed the opinion [25] that 'Pfennig is highly relevant to the effect of ss 97, 98 and 101 of the Evidence Act, in that the principles there stated, concerning circumstances in which the probative force of similar fact evidence substantially outweighs its prejudicial effect are directly applicable to questions raised for decision by ss 101 and 137 of the Evidence Act: R v AH (1997) 42 NSWLR 702'. Then, after referring to what had been said in Pfennig by Mason CJ, Deane and Dawson JJ at 482 ‑ 483 concerning the test to be applied for the admission of propensity evidence, he said, of that passage [29]:
In my opinion, what it must mean is that, if it first be assumed that all the other evidence in the case left the jury with a reasonable doubt about the guilt of the accused, the propensity evidence must be such that, when it is considered along with the other evidence, there will then be no reasonable view that is consistent with the innocence of the accused. That is, the propensity evidence must be such that, when it is added to the other evidence, it would eliminate any reasonable doubt which might be left by the other evidence.

Effect of s 31A

Returning to the legislation in this State, it has been accepted that s 31A substantially altered the common law. It abrogated the test that propensity evidence is inadmissible if there is a rational view of it, when considered with other relevant evidence, that is inconsistent with the guilt of the accused: Wood v The State of Western Australia [2005] WASCA 179; Donaldson v The State of Western Australia [2005] WASCA 196; (2005) 31 WAR 122 [102] ‑ [130]; Di Lena v The State of Western Australia [2006] WASCA 162; (2006) 165 A Crim R 482 [44] ‑ [73]; Noto v The State of Western Australia [2006] WASCA 278; (2006) 168 A Crim R 457 [26]. Instead, the evidence is admissible if the court considers that each of the requirements of subs (2)(a) and (b) have been satisfied.

Significant probative value: s 31A(2)(a)

The evidence in question must obviously be relevant before it can be admitted into evidence. That is to say, it must be such as could rationally affect, directly or indirectly, the assessment of the probability of the existence of a fact in issue in the proceeding: Goldsmith v Sandilands [2002] HCA 31; (2002) 76 ALJR 1024 [2] (Gleeson CJ); Phillips v The Queen [2006] HCA 4; (2006) 225 CLR 303 [50]. It could otherwise have no probative value, let alone 'significant' probative value.
Before evidence can have significant probative value it must be such as 'could rationally affect the assessment of the probability of the relevant fact in issue to a significant extent: ie, more is required than mere … relevance': Zaknic Pty Ltd v Svelte Corporation Pty Ltd (1995) 61 FCR 171, 175 ‑ 176. Heydon [21245] suggests that significant probative value is something more than mere relevance but something less than a 'substantial' degree of relevance and that it is a probative value which is 'important' or 'of consequence'. He makes the point that the significance of the probative value of tendency evidence must depend on the nature of the facts in issue to which it is relevant and the significance or importance which that evidence may have in establishing the fact: Lockyer (1996) 89 A Crim R 457, 459; Lock (1997) 91 A Crim R 356, 360 ‑ 361; Fordham (1997) 98 A Crim R 359, 370; Jacara Pty Ltd v Perpetual Trustees WA Ltd [2000] FCA 1886; (2000) 106 FCR 51 [77] ‑ [82]; The State of Western Australia v Osborne [2007] WASCA 183 [13].

The s 31A(2)(b) comparison

Once the evidence is found to have significant probative value, either by itself or taken with other evidence, the court must engage in the process contemplated by s 31A(2)(b). Because there will already have been an assessment of the probative value of the evidence (taking into account the purpose for which it is adduced and its likely effect when considered together with the other evidence), it is necessary, next, to assess the degree of risk of unfairness in the trial that will be brought about by the admission of the evidence.
Geoffrey Flatman QC and Dr Mirko Bagaric, 'Non‑similar Fact Propensity Evidence: Admissibility, Dangers and Jury Directions' (2001) 75 Australian Law Journal 190, 199, identify a threefold prejudice (only part of which is present in this case) involved in the introduction of similar fact evidence, as follows:
1.'The over strong tendency to believe that the defendant is guilty of the charge merely because he is a likely person to do such acts' (Wigmore, Evidence (1940), p 650; see also Waight and Williams, [Evidence: Commentary and Materials (5th ed, LBC, 1998)], p 426 (citing R v Bailey [1924] 2 KB 300 at 305: 'it is easy to derive from a series of unsatisfactory allegations … an accusation which at least appears satisfactory … to collect from a mass of ingredients, not one of which is sufficient, a totality which will appear to contain what is missing')). This comes down to the belief that the jury will over‑estimate the cogency of the similar fact evidence; the jury may act illogically by giving too much weight to the evidence. In BRS v The Queen ((1997) 191 CLR 275) Kirby J stated:
'research confirms the common tendency to infer from particular conduct character traits which are then used to justify predictions and estimates about other conduct. However, objectively, such predictions are frequently shown to be unwarranted (Ibid at 322).'
2.'The tendency to condemn, not because he is believed guilty of the present charge, but because he has escaped punishment from other offences' (Wigmore … p 650). Thus, 'there might be a tendency for the jury to punish the accused for past misconduct by finding the accused guilty of the offence charged' (R v Rarru (1996) 107 CCC (3d) 82). This second danger refers not to a possible defect in logic that may be triggered by similar fact evidence, but the fear that the jury will convict solely due to a bias against the accused: 'sentiments of revulsion and condemnation … might well deflect [the jury] from the rational dispassionate analysis upon which the criminal process should rest' (Ibid). In short, it is thought that the reception of similar fact evidence may result in a bias being formed against the accused which will taint the jury's decision.
3.'The jury might become confused [or distracted] as it concentrates on resolving whether the accused actually committed the similar acts' (Ibid; see also … Pfenning v The Queen (1995) 182 CLR 461 at 512).
(See also Heydon [21135] ‑ [21150].)
When assessing the risk of an unfair trial for the purposes of this hypothetical comparison, the court will take into account any directions that might be given to the jury in an attempt to overcome the prejudice, and their likely effect on the jury. It is important to bear in mind in this respect that, when propensity evidence is admissible as such because it meets the requirements for the admission of evidence of that kind, a standard propensity warning will not be required: Noto [27]. In KRM v The Queen [2001] HCA 11; (2001) 206 CLR 221, McHugh J said (235):
If evidence tendered to prove a subsidiary issue (including the relationship between the parties) reveals the criminal or discreditable conduct of the accused, the judge will often, but not always, have to give a propensity warning … And if evidence has been admitted generally as propensity evidence, it is difficult to see how a propensity direction is ever required. In that class of case, the evidence is tendered to prove that the accused is the type of person who is likely to have committed the crime with which he or she is charged. To require a propensity direction would contradict the basis on which the propensity evidence is admitted. And that is so, whether the propensity evidence consists of uncharged acts or evidence supporting the charge in one count that is also relevant to charges in other counts in the presentment. (emphasis in original)
Of course, that does not necessarily mean that, in a case in which propensity evidence is admissible as such, no warning of any kind is required. Whether a warning is required and, if so, the terms of that warning, will depend upon the circumstances of the individual case. It is important to have regard for the purpose for which the evidence is admitted and to tailor the directions accordingly: Gipp v The Queen [1998] HCA 21; (1998) 194 CLR 106 [77] (McHugh & Hayne JJ). It is also important to bear in mind what was said by Kirby J in BRS (330) as follows:
It is … necessary to remind oneself that a judge's direction to a jury must be considered in the context of the entire trial and in the light of the issues litigated and the submissions made. This Court should exercise restraint before adding to the catalogue of warnings required of judges conducting criminal trials. If overlooked, or considered unnecessary in a particular trial, they may simply give rise to unmeritorious appeals. But, in certain fundamental matters, the judge carries an unavoidable obligation to provide directions which are necessary to prevent a perceptible risk of a miscarriage of justice. The risk must be a real, and not a fanciful one to attract the intervention of an appellate court. It is not enough that a conclusion is reached that a particular warning would have been desirable. It must be shown to have been necessary in order to avoid the risk of a miscarriage of justice. (footnotes omitted)
Having identified the probative value of the evidence and the degree of risk of an unfair trial, the court must turn its attention to the conclusion that fair‑minded people would draw from a comparison of the two. These fair‑minded people are presumably reasonable members of the general public who are not lawyers: Raybos Australia Pty Ltd v Tectran Corporation Pty Ltd [No 9] (Unreported, NSWCA, 27 November 1990) (20), cited in Australian National Industries Ltd v Spedley SecuritiesLtd (in liq) (1992) 26 NSWLR 411, 419; Johnson v Johnson [2000] HCA 48; (2000) 201 CLR 488 [53] (Kirby J). However, the legislature must be taken to have assumed that such people would have informed themselves of 'at least the most basic considerations relevant to arriving at a conclusion founded on a fair understanding of all the relevant circumstances': Johnson [53] (Kirby J).
The comparison that these fair‑minded people are to be assumed to have undertaken is problematic. As McHugh J pointed out in Pfennig (528), 'prejudicial effect [or, I would suggest, the degree of risk of an unfair trial] and probative value are incommensurables' that have 'no standard of comparison'. It is not easy to compare the probative value of the evidence with the degree of risk to the fairness of the trial that is brought about by its introduction. That risk arises because the jury might use the evidence in an impermissible way. In a case in which the evidence is led for its propensity value (for example, as identification evidence), the risk of unfairness might increase proportionately with the probative value of the propensity evidence. In effect, the test provided by subs 2(b) requires an assessment to be made whether fair‑minded people would think that the interests of justice require the admission of the evidence despite the risk: cp the similar comments made by McHugh J in Pfennig (528) in respect of his statement of the position under the common law.

The present case

That brings me back to the present case. The propensity evidence (which is cogent, the appellant having undoubtedly been convicted of the other offences) was said to have been relevant, essentially for two reasons. The first is that it lent support to the proposition that the person identified by witnesses as being 'similar' to the offender was in fact the offender. Counsel for the respondent contends in this respect that, because the appellant has a tendency to steal cars and violently resist arrest, this places him in a class of people who are more likely than others to have committed an offence of the kind charged. That, in turn, was said to have added strength to the tentative identification of him by a number of witnesses. Secondly, the propensity evidence was said to support the evidence of Mr Strachan. Counsel for the appellant contended that, because the propensity evidence added to the likelihood that the appellant had committed the offence charged, it was more likely that Mr Strachan had correctly identified the appellant, that the appellant had confessed to the commission of the offence and that the appellant's confession was genuine.
It is not in dispute that the evidence of the appellant's other criminal conduct is propensity evidence as defined by s 31A(1). Nor is it in dispute that the evidence was relevant for the first of the reasons offered by counsel for the respondent. Whether it had significant probative value for that reason, or for any other reason, is a more difficult question.
Had the evidence stood on its own, I very much doubt that it could, to any significant extent, rationally have affected the assessment of the probability that the appellant was the man who had broken into Mr Yap's car and subsequently stabbed Constable Giocas. Experience reveals that there are many criminals who steal cars and subsequently attempt to evade arrest, if pursued. The fact that the appellant was one of these (and it is not suggested that there is anything unique about his modus operandi) did little to establish that he committed the present offence. As I have said, what it did was include him in a class of persons who were more likely than others to have committed an offence of this kind. However, given that the class very probably encompasses a large number of people, that, of itself, could not carry a great deal of weight.
The fact that the appellant, when challenged, had a tendency to resort to violence made it more likely that he was the offender. However, the class of car thieves who are prepared to use violence to evade arrest, while smaller than the first class to which I have referred, may nonetheless consist of a sufficient number of people to render that evidence, taken on its own, of limited value. Importantly, the violent means used by the appellant had never encompassed a stabbing, or even the infliction of grievous bodily harm, and, although he had previously brandished a knife, this had happened on only one occasion.
Nevertheless, it was significant, in the context of this case, that the witnesses had picked out (albeit tentatively) a man who belonged to the smaller class to which I have referred. The fact that the appellant was a member of this class made it more likely that these witnesses had picked out the right man. It seems to me that the propensity evidence consequently had significant probative value.
That brings me to the degree of risk of an unfair trial. There is no doubt that the evidence of the commission of other offences by the appellant had a high prejudicial value. Although admissible for propensity purposes, there was a risk that the jury would over‑estimate the value of the evidence and reason that, because the appellant had committed offences of a somewhat similar kind in the past, he must have been the person who committed the present offence, more especially in circumstances in which there was evidence that he looked like the offender. Alternatively, they may have thought that, because the appellant, who looked like the offender, was a member of a class of persons likely to commit crimes of this kind, they were relieved of their obligation to closely examine the other evidence against him. Further, the evidence of other criminal conduct may have caused them to be biased against the appellant.
Next, it is necessary to identify the factors that must be taken into account in a comparison of the probative value of the evidence with the risk of unfairness.
Although I accept that the propensity evidence lent significant support to the evidence of those who had been unable to make any positive identification of the appellant, I am unable to accept that it lent substantial support to that evidence. Even though the witnesses had tentatively identified the appellant as having a similar appearance or similar characteristics to the offender, that fact (weakened by the inaccurate and inconsistent descriptions given by the witnesses), taken together with his membership of the smaller class to which I have referred, fell well short of establishing his guilt. The witnesses still could not say that he was the man that they had seen. There may well have been others in the class, or a new member of the class, who looked like the offender.
Next, the propensity evidence did not lend any substantial support to the evidence of Mr Strachan. He was firm in his positive identification of the appellant. He was unequivocal in his evidence of the confession made by the appellant. These critical aspects of his evidence (which seem to me to have been likely to stand or fall together in the assessment of the jury) were either credible and reliable or they were not. All that the propensity evidence did was add a factor, albeit a significant one, that made it more likely that the appellant was the offender.
If the evidence of Mr Strachan was reliable, it seems inconceivable that the appellant would not have been convicted, even without the propensity evidence. If, on the other hand, Mr Strachan was regarded by the jury as being dishonest or unreliable, the case against the appellant was not compelling, even with the propensity offence. What the jury would then be left with was the very tentative (and in some respects contradictory) evidence of similarity in appearance or characteristics to the offender (in some cases, arising out of less than ideal identification procedures), the evidence of the appellant's membership of the more limited (but not especially limited) criminal class to which I have referred and the somewhat equivocal evidence of the telephone conversations. Against that, there was no fingerprint or DNA evidence connecting the appellant with the attempted car theft or the stabbing.
All of this had to be considered together with the risks to which I have referred that would arise out of the introduction of the propensity evidence. In my opinion, fair‑minded people would undoubtedly consider that impermissible reasoning of the kind that I have earlier identified would result in an unfair trial. They would also conclude that, if Mr Strachan's evidence was regarded by the jury as credible and reliable, the disputed evidence was unnecessary. Next, they would consider that, if his evidence was not so regarded (and, as I have said, in my opinion any finding of unreliability concerning his evidence of the confession would necessarily have extended to his positive identification of the appellant), the propensity evidence was probably insufficient to establish the appellant's guilt, when considered together with all of the other evidence. Those conclusions would, in turn, drive these fair‑minded people to the conclusion that, unless the potential unfairness could be adequately guarded against by a direction from the trial judge, the public interest in adducing the evidence should not have priority over the risk of unfairness.
It is unlikely that the risk of unfairness could adequately have been guarded against by a direction in this case. This was a case in which, as I have stressed, the evidence of the appellant's commission of other offences was admitted solely for its propensity value, notwithstanding that it did no more than place the appellant in a class of potential offenders. The prejudicial value of that evidence is so great that it is difficult to be satisfied that even the strongest of warnings would be enough to overcome it. It is precisely because of doubt concerning the ability of a warning to overcome the risk to the fairness of the trial process brought about by the admission of propensity evidence that the courts have arrived at, and repeatedly applied, a stringent test for the admission of evidence of that kind. It consequently seems to me that informed fair‑minded people would conclude that, even allowing for the giving of a suitable warning, the degree of risk was so high that the public interest in adducing all relevant evidence of guilt could not have priority over that risk.

I should add, with due respect, that the directions that were given by the trial judge in this regard were confusing.
As Miller JA points out, the trial judge repeatedly warned the jury, rightly, that the fact that the appellant had previously stolen cars and, when confronted, resisted arrest, did not automatically mean that he was guilty of the offence with which he was charged. He also told the jury that the other offences were not 'similar in all details' to the present offence. He said (ts 636):
You have heard lots of things about what [the appellant] has done in the past and things like that. You are not concerned about those matters at all, don't be distracted by what he may have done in the past in other offences he may have committed. Your focus is, did he commit this offence on this day and that's what your focus is. The other matters, I will tell you what use those others matters are and about previous conduct, but just because you have heard about that doesn't mean that he is the sort of person who is likely to have committed this offence because he did those things in the past. It is part of the material that you have been given, but focus upon that charge which is before you on that indictment, that's what the trial is about, not what he may have done in the past.
That direction, whilst of assistance to the appellant, was in error to the extent that the trial judge said that the fact that the other offences had been committed did not mean that the appellant was 'the sort of person who is likely to have committed this offence'. Evidence of the other offences was adduced for the very reason that it established that the appellant was the sort of person who was likely to have committed an offence of the kind charged.
When he returned to this issue later in his charge, the trial judge said (ts 644):
The State have also put before you evidence in relation to other incidents in which [the appellant] was involved. Some, two of them, one in 1998, one in 2002 and one not long after the incident on 3 April 2003. The State is not saying that they were identical incidents but what the State is saying, that they are of a similar nature, that there has been a car theft involved and there has been violence in one form or another offered by [the appellant] to try and escape apprehension when he has been confronted and people have tried to arrest him.
So that evidence along with the evidence of the eyewitnesses, although it's not directly - they are not saying they are absolutely certain it was [the appellant], is put before you by the State to consider and to support Mr Strachan's evidence as being inherently believable even though there may be some difficulties with it.
That direction is, with respect, problematical. It is difficult to see how evidence of other criminal conduct could support Mr Strachan's evidence 'as being inherently believable', whether there were or were not difficulties with it. The fact that the appellant was a member of a criminal class with tendencies to steal cars and resort to violence in order to escape arrest added nothing to the question of whether the whole of Mr Strachan's evidence was or was not 'inherently' credible. All it did was to add to the likelihood that the appellant had committed the offence charged.
The trial judge returned to the issue still later in his charge. He said (correctly) that the jury could not conclude that the appellant was guilty merely because he had committed similar offences on other occasions. However, he also said (ts 670) that the State was relying upon the evidence of other offences 'to bolster Mr Strachan's evidence'. He did not explain what he meant by that, leaving it open to the jury, or some of them, to think that this was another reference to the jury's ability to treat that evidence as supporting Mr Strachan's evidence 'as being inherently believable'.
The trial judge also said (ts 670):
Whether or not the two previous offences showed that he had this propensity or tendency to commit this type of offence is for you to decide as is the offence which occurred after the commission of the offence on 3 April … You are entitled, if you find that there is this tendency or propensity, to take [it] into account when determining whether or not [the appellant] is guilty of this offence.
Then, towards the end of his charge to the jury (ts 674), the trial judge said:
Why you have heard about the evidence of the previous convictions is to let you make some assessment of the character of [the appellant]. You know that he doesn't come before you with a blameless character, but you can't reason because he has committed offences in the past he must have committed this offence. That's not what it's there for, it's just to let you get a fuller picture of him, but you can't assume that he's guilty of this offence simply because he has been convicted prior to 3 April and subsequently to the 3 April … [Those offences] are not relevant at all to the likelihood of him having committed this particular offence, it's likely [sic] simply for you to make a judgment about his character.
After the completion of the trial judge's charge, the then counsel for the parties questioned the suggestion, in this last passage, that the evidence could be used for the purposes of making a judgment about the appellant's character. The trial judge consequently agreed to re‑direct on that point. When the jury were recalled, he said, in this respect (ts 684):
Just going back again to Mr Strachan's evidence and the way that the state relies upon the evidence of people who recognised [the appellant] and the telephone calls and the previous incidents, those were led for the purpose of establishing the objective improbability of [the appellant] with the history that you heard about of the other offences and the identification - not the identification but the picking out of photographs of him as being the person involved at the Cultural Centre who is alleged to have done the things that the state alleges he has done as being chosen as the offender when in fact he wasn't the offender, so it's to bolster Mr Strachan's evidence. It's not to be used in any way to look at [the appellant] as a person being of such a character that he is likely to have committed these offences.
The State are not relying upon it for that purpose at all. It's simply to lead the evidence to establish the improbability of the person involved who did those things not being the person Mr Strachan was talking about.
With respect, this direction seems to me also to be confusing. As I have stressed, the evidence was led because it established the appellant's membership of a class of persons likely to commit an offence of this kind, thereby adding to the likelihood that he had in fact committed the offence. While this evidence added support to the prosecution case, including Mr Strachan's evidence, it did not, in any other sense, 'bolster' that evidence and this reference, too, may have misled the jury.
Also, the trial judge was wrong to say, as he did, that the evidence was not to be used so as to establish that the appellant was a person of such a character that he is likely to have committed the offences in question. It was to be used for that purpose. That was the reason why the State had adduced it. That error, taken with the concluding sentence of the re‑direction, to the effect that the evidence had been led 'to establish the improbability of the person involved … not being the person Mr Strachan was talking about', would have left the jury uncertain of the use to which the evidence should be put. Once again, they may well have concluded that, in some way or another, the evidence was admissible for the purpose of bolstering Mr Strachan's credibility, or overcoming the difficulties with his evidence to which the trial judge had referred.
It consequently seems to me that grounds 4 and 7 have been made out.
Once the conclusion is reached that the evidence should not have been admitted, and that its prejudicial value was not overcome by the trial judge's directions, there can be no basis for the application of s 30(4) of the Criminal Appeals Act 2004 (WA). In circumstances in which there was the prospect of impermissible reasoning by the jury, and in which the prejudicial value of that reasoning was very high, it seems to me to be impossible to say that no substantial miscarriage of justice has occurred. As I have said, the case against the appellant was strong only if the jury found that the evidence of Mr Strachan was reliable. There is no way of knowing to what extent the jury found it to be so, or to what extent they would have done so had the propensity evidence not been available. Nor is there any way of knowing whether the jury understood that the evidence could not be used for the purpose of bolstering Mr Strachan's overall credibility. It seems to me to be reasonably possible that some of them might have understood the trial judge to be saying that it could be used for that purpose, more especially in circumstances in which he had, at one point, told them that the evidence could 'support Mr Strachan's evidence as being inherently believable'. In these circumstances, it is impossible to assess the impact of the irregularities on the fairness of the trial and hence to be affirmatively satisfied that no substantial miscarriage of justice has occurred: Libke v The Queen [2007] HCA 30; (2007) 235 ALR 517 [46] (Kirby & Callinan JJ, dissenting in the result).
It follows that the appellant's conviction should be quashed and a re‑trial ordered.

Ground 5 of the appeal against conviction

I agree with Miller JA, for the reasons that he has given, that ground 5 has not been made out.

Ground 6 of the appeal against conviction

Ground 6 contends that the trial judge erred in directing the jury that the evidence of the eye witnesses was capable of supporting 'Mr Strachan's evidence as being inherently believable' and as being able to 'bolster Mr Strachan's evidence'. The context in which these comments were made by the trial judge has been set out in the judgment of Miller JA. I will not repeat what he has said in that respect.
I have already said that Mr Strachan's evidence was either inherently truthful and reliable or it was not. The weak 'identification' evidence added to his evidence only in the sense of establishing that, so far as most of the witnesses were concerned, the offender was in some respects similar in appearance to the appellant and, in the case of one of them, that he had an unusual gait. That evidence added to the likelihood that the appellant was the offender, just as the evidence of his convictions for other offences did, but it did not make Mr Strachan's evidence 'inherently believable' or 'bolster' his evidence.
Although it is unnecessary to decide this ground, having regard for the conclusion at which I have arrived in respect of grounds 4 and 7, in my respectful opinion the error made by the trial judge in this respect supports the conclusion that the verdict of the jury should not stand.

Proposed ground 8

I agree with Miller JA, for the reasons that he has given, that leave to appeal on this proposed ground should be refused.

The appeal against sentence

Given that I would set aside the conviction, it is unnecessary for me to deal with the appeal against sentence. However, I agree with the conclusion arrived at by Miller JA in that respect.
This offence was serious. The fact that it involved the stabbing of a police officer who was attempting to apprehend the appellant made it even more serious. The courts have an obligation to endeavour to protect police officers, when acting in the proper performance of their duty, by doing what they reasonably can to deter those who might otherwise assault them. Consequently, although a sentence of 8 years' imprisonment (12 years prior to the mandatory reduction required by the transitional provisions in sch 1, cl 2(1) of the Sentencing Legislation Amendment and Repeal Act 2003 (WA)) is very severe, I would not have interfered with it had it stood on its own. However, I respectfully agree with Miller JA that

the sentencing judge made no, or no sufficient, allowance for the operation of the totality principle (as to which see Postiglione v The Queen [1997] HCA 26; (1997) 189 CLR 295, 307 ‑ 308 (McHugh J)), with the result that his discretion miscarried.

I also agree with Miller JA that a sentence of 6 years' imprisonment without eligibility for parole (9 years' imprisonment before taking into account the transitional provisions), to be served with effect from 26 August 2006, is appropriate, when added to the term of 3 years and 4 months' imprisonment already served by the appellant (making up a total of 9 years and 4 months' imprisonment that will have been served by the appellant before being released).

Conclusion

It follows that I would allow the appeal against conviction, quash the conviction and order a re‑trial. Were it necessary for me to address the appeal against sentence, I would have allowed it for the reasons and with the outcome suggested by Miller JA.
MILLER JA: The appellant was charged on indictment that, on 3 April 2003, at Perth, with intent to maim, disfigure, disable or do grievous bodily harm to Normie Philip Giocas, he unlawfully did grievous bodily harm to Normie Philip Giocas. The offence is one that carries a maximum sentence of imprisonment of 20 years.
The appellant was tried before Healy DCJ and a jury in the District Court at Perth and, on 27 February 2007, was convicted as charged. He was sentenced on 28 February 2007 to 8 years' imprisonment, backdated to 26 August 2006. The sentencing judge declined to order eligibility for parole.
The appellant appeals against both conviction and sentence.

Appeal against conviction

There are eight grounds of appeal against conviction. Leave to appeal on grounds 1 to 7 inclusive was granted by Wheeler JA on 2 October 2007. The application for leave to appeal on ground 8 was referred to the Court of Appeal to be heard together with the appeal.
The grounds of appeal against conviction (deleting particulars in relation to grounds 5 and 6) are as follows:

1.The admission of evidence, of Giocas as to the identification parade held at Hakea Prison and the Police Station on the 9th May as circumstantial evidence has resulted in a miscarriage of justice.
2.The admission of evidence of Symington as to the identification parade held at Hakea Prison on the 22nd day of May 2003 as circumstantial evidence has resulted in a miscarriage of justice.
3.The admission of evidence of Woollard as per the photoboard identification taken on the 22nd April 2003 as circumstantial evidence has resulted in a miscarriage of justice.
4.His Honour erred in admitting conduct of the appellant's past convictions under Section 31A of the Evidence Act in that
(a)They did not have significant probative value
and
(b)It cannot be said that the probative value of the evidence compared to the degree of risk of an unfair trial is such that fair minded people would think that the public interest of adducing all relevant evidence of guilt must have priority over the risk of an unfair trial.

5.His Honour erred in his direction to the jury on the use of the circumstantial evidence used in relation to the identification procedures of Giocas, Symington and Woollard, in that it was said the evidence did not identify the accused but then proceeded to direct in terms that did.
6.His Honour erred in directing that the evidence of eye witnesses as capable of supporting Mr Strachan's evidence as being inherently believable and bolstering that evidence when such evidence was not capable of this ...
7.His Honour failed to adequately direct on the circumstantial evidence of the past criminal conduct and such failure has led to a miscarriage of justice in that
PARTICULARS
1.His Honour spoke of the evidence as being capable of supporting Mr Strachan's evidence as being inherently believable - when it was not.
2.His Honour referred to the evidence as being capable of bolstering Mr Strachan' s evidence - when it was not.
3.His Honour failed to point that the evidence led of past criminal conduct on its own does not establish anything and does not establish that it was Mr Dair who injured the complainant.
4.His Honour incorrectly charged that the jury was entitled to find that there is a tendency or propensity to take into account when determining whether or not Mr Dair is guilty of the offence.
5.Advised the jury could take into account the likelihood of him having committed this particular offence based on his character even though it was corrected in a redirection.

8.It is submitted that whilst Strachan could give evidence of his knowledge of the appellant and features and how the appellant may have walked he can not purport to usurp the jury on the question of identification and the question as to whether the surveillance and identikit picture is the appellant. See Smith v R [2001] HCA; 206 CLR 650

The evidence at trial

To understand the grounds of appeal, it is necessary to summarise the prosecution case. (The appellant did not testify at trial.)
The offence was alleged to have occurred on 3 April 2003 in the Perth City Library carpark in Francis Street, Northbridge. At about 3.45 pm that day, Thiam Hoi Yap (Mr Yap) went to his vehicle, which was parked in the library carpark. He found the driver's door open and a man sitting in the driver's seat. He yelled at the man and the latter grabbed a bag from the passenger's side of the vehicle and jumped out. He ran to an exit stairwell. Mr Yap ran after him, but the person he was chasing (the offender) turned towards him, put up his hands, waving something as if to ward him off. Mr Yap could see what he thought was black metal shining in the offender's hand. He put up his own hands to protect himself and ceased pursuit. The offender then ran off.
The complainant (Giocas) was an off‑duty police officer who was in the carpark at the time. He observed what was happening and, after speaking briefly to Mr Yap, he chased the offender. He pursued him downstairs towards the lower level of the carpark and called out to him, 'Stop, police'. He said that he flashed his police identification. The offender made no attempt to stop, despite repeated calls upon him to do so. He ran up a ramp, through the carpark foyer and into the plaza area of the Perth Cultural Centre.
Giocas rang police communications to report what was happening. He then chased the offender into the open area of the plaza. The offender stopped running, turned and faced Giocas. He took some steps towards him, called out, 'Come on, cunt' and swung something at him with his hand. Giocas felt something strike his neck. He did not immediately experience pain, but he knew he had been hit in the neck. He backed away. He did not immediately feel blood, but shortly afterwards, he saw blood flowing down his shirt. He then observed his hand to be covered in blood. He was forced to let the offender go.
The offender escaped, leaving Giocas awaiting an ambulance. He required urgent medical assistance. He had suffered a partial severance of the superior thyroid artery. The injury was such that it endangered his life. Substantial surgery at Royal Perth Hospital repaired the damage to Giocas' neck. He made a full recovery.
Police searched the scene. They located a screwdriver in the stairwell of the carpark. This was the stairwell through which the offender had run whilst being pursued. The screwdriver was later compared with damage to Mr Yap's car. The comparison suggested to a police officer that marks on the car were consistent with the use of that screwdriver.

The issue of identification

It was further argued by counsel for the respondent on the appeal that this propensity evidence was probative, because it bolstered the evidence of the witness Strachan of the admission which he claimed had been made to him by Dair in their private meeting.
The reception of this evidence and the manner in which the learned trial judge, Healy DCJ, directed in respect of it, are the subject of the fourth and seventh grounds of appeal. Essentially, the appellant complains that the evidence should not have been received because its prejudicial effect exceeded its probative value and that even when it was received the learned trial judge failed to identify, or adequately identify, the only permissible manner in which it could be utilised.
Propensity evidence, like all evidence, to be admissible must be relevant and material in the sense that it may, if accepted, assist in the proof of the offence charged or some particular element of that offence. Its capacity to contribute logically to the proof of the offence or some element of the offence is what makes it probative. If it does not have that capacity, that is if it is not probative, it is not relevant and is therefore inadmissible. The difficulties encountered with propensity evidence, however, derive from its dual or multiple effects. Apart from being probative of the fact or facts in issue in the prosecution before the court, the propensity evidence will usually involve the disclosure of some unlawful or discreditable conduct by the accused on previous occasions. Therefore it has the real potential of prejudicing the accused in the eyes of the jury and can lead to the effect of generating a process of reasoning that because the accused broke the law, or was engaged in some discreditable conduct on one or more previous occasions, he is for that reason more likely to have committed the offence with which he is charged. It is to avoid that often specious form of reasoning that the controls contained in the principles developed by the courts have been fashioned. Usually that trend of reasoning is impermissible but there are sometimes occasions where, because of its potential cogency, evidence leading to this form of reasoning may exceptionally be admissible because the evidence of disposition or propensity is logically probative in the particular case (see generally, Ligertwood: Australian Evidence (2004) [3.23] ‑ [3.29]).
This is more easily expressed in principle than applied in practice. The approach is that the fact that the accused has committed an offence or offences or behaved discreditably on previous occasions is usually presumed to provide no logical foundation for concluding that he is guilty of the offence with which he presently stands charged. The exception to this approach occurs where there is some particular similarity, feature or hallmark of the offence presently alleged to have been committed, or the way in which it was committed, which renders it likely that the offence with which the accused was charged was committed by the same person who committed the previous offences or engaged in that earlier conduct. This is said to require some 'striking similarity', some 'hallmark', some 'unique feature', some 'stamp of identity' or some other talismanic feature of the offence charged and the earlier conduct involving the accused which suggests that it is unlikely to be coincidental that the offence charged would have been committed by someone other than the person proved to have committed the other offences or to have engaged in the other discreditable conduct of this accused on the earlier occasions. Even then however, mere propensity or dispositional evidence may be probative if it provides some 'underlying unity' to explain a series of events - contrast DPP v Boardman [1975] AC 421 at 429 and 444, and Thompson v The Queen (1989) 169 CLR 1 per Deane J at 32 and Gaudron J at 39 ‑ 40.
Accordingly, the first step in dealing with any issue concerning the admissibility of propensity evidence is to determine whether it is capable of being probative of the offence charged. If it is, the second stage of dealing with the admissibility of the evidence must be undertaken, namely, to determine whether, in the circumstances of the case in question, the probative effect of the evidence justifies the reception of the evidence notwithstanding its prejudicial tendency. It is here that the legislature has prescribed, by s 31A of the Evidence Act, a particular standard to be applied when conducting this balancing act.
The third stage of the exercise must be performed if it is decided, on balance, that the probative effect of the evidence justifies its admissibility because then the jury will need to be directed that it can only attend to the probative effect of the evidence and must not simply adopt the process of reasoning based on prejudice, namely that lawbreaking on a previous occasion renders it more likely that this accused will have committed the offence for which he is on trial. Sometimes this is a very fine distinction because the probative force of the evidence may lie entirely in its dispositional tendency.
In Pfennig v The Queen [1995] HCA 7; (1995) 182 CLR 461, Mason CJ, Deane and Dawson JJ said (481):
It was also accepted that, in order to be admissible, propensity evidence must possess 'a strong degree of probative force' (Markby (1978) 140 CLR at 117; Perry (1982) 150 CLR at 586, 589, 604 and Sutton (1984) 152 CLR at 533) or the probative force of the evidence must clearly transcend the prejudicial effect of mere criminality or propensity. Very often, propensity evidence is received when there is a striking similarity between different offences or between the evidence of different witnesses. In particular, it was recognised that the existence of such striking similarity is necessary in cases such as Sutton where the prosecution seeks to lead the evidence on the basis that the similarity between different offences founds a conclusion that they must have been committed by the one person with the consequence that evidence which would be admissible to show that an accused committed one of the offences is admissible to prove that he or she committed another or the others of them.
and later, in the same judgment, their Honours said (482 ‑ 483):
Because propensity evidence is a special class of circumstantial evidence, its probative force is to be gauged in the light of its character as such. But because it has prejudicial capacity of a high order, the trial judge must apply the same test as a jury must apply in dealing with circumstantial evidence and ask whether there is a rational view of the evidence that is consistent with the innocence of the accused (Hoch (1988) 165 CLR at 296). Here 'rational' must be taken to mean 'reasonable' (see Peacock v The King (1911) 13 CLR 619 at 634 and Plomp v The Queen (1963) 110 CLR 234 at 252) and the trial judge must ask himself or herself the question in the context of the prosecution case; that is to say, he or she must regard the evidence as a step in the proof of that case. Only if there is no such view can one safely conclude that the probative force of the evidence outweighs its prejudicial effect. And, unless the tension between probative force and prejudicial effect is governed by such a principle, striking the balance will continue to resemble the exercise of a discretion rather than the application of a principle.
Also in Pfennig, Toohey J, who delivered separate but concurring reasons, said (507):
The judgment of Mason CJ, Wilson and Gaudron JJ in Hoch identifies the basis upon which evidence of similar facts is admitted. 'The criterion of its admissibility is the strength of its probative force.' ((1988) 165 CLR at 294). However, it remains with the trial judge to consider the prejudicial effect of the evidence and whether, in the circumstances, it is just to admit it. That is not precisely the same as saying that the evidence is admissible if its probative force exceeds its prejudicial effect. The probative force of the evidence, in accordance with the criterion discussed, is the basis for its admissibility; nevertheless, the trial judge may exclude it. Evidence that an accused has committed other relevant offences must inevitably have a prejudicial effect. But, in the language of Director of Public Prosecutions v P, it may nevertheless be 'just' to admit the evidence. The reference to just aptly conveys the notion that it is not only the interests of the accused that are involved. The legitimate interests of the Crown and of the community cannot be overlooked.
In Pfennig (475), Mason CJ, Deane and Dawson JJ also cited with approval the observations of Gibbs CJ in Sutton v The Queen (1984) 152 CLR 528 at 535 about whether there is a 'striking similarity' between the similar facts and the fact in issue.
Section 31A(2) of the Evidence Act has varied the pre‑existing law on this subject. For the propensity evidence to be admissible, it must 'have significant probative value' thereby emphasising that a merely tenuous, oblique or remote probative potential of the evidence would be insufficient to justify admission. This is in accord with Pfennig (481). When coming to the balancing exercise, which I have described as being the second stage of the process to be followed when dealing with a contested issue about admissibility of propensity evidence, s 31A(2)(b) strikes the balance between recognised probative value and potential prejudicial effect without ignoring the risks of the propensity evidence to lead, if its use is not properly explained by a trial judge to the jury, to impermissible and illogical reasoning. The statute achieves this effect by its very language which, so far as is material provides:
31A(2)Propensity evidence … is admissible in proceedings for an offence if the court considers -
(a)…
(b)that the probative value of the evidence compared to the degree of risk of an unfair trial, is such that fair‑minded people would think that the public interest in adducing all relevant evidence of guilt must have priority over the risk of an unfair trial.
which, obviously, is very similar to the principle explained by Toohey J in Pfennig.
Turning to the question whether the propensity evidence in the present case was rightly or wrongly admitted requires, first, a determination of whether or not it was probative of the fact in issue. The fact in issue was the identity of the offender who stabbed Constable Giocas. The other evidence relied on by the prosecution to prove the identity of the appellant as the offender was the evidence of Mr Strachan comprising, as it did, his evidence of recognition of the person shown on the CCTV videotape of the attack as the appellant, based on his knowledge and acquaintance of the appellant and his recognition of the appellant's distinctive gait, together with Strachan's evidence of a subsequent conversation between himself and the appellant in which, according to Strachan, the appellant admitted having stabbed the police officer on that particular occasion.
Then there was the circumstantial identification evidence from the other witnesses, the owner of the car which the offender had broken into; the lady in the carpark who had seen an altercation between Mr Yap and the offender; Constable Giocas himself; and the young woman who had seen the assault on the plaza outside the parking area. None of these persons was able to give evidence of positive identification of the appellant but each, in various ways, described the appearance of the offender and three of them, Constable Giocas, the lady in the carpark and the young woman in the plaza outside, when shown photoboards or when observing a group of other men at Hakea Prison, tentatively ventured the opinion that the appellant, in various ways, resembled the offender although none was able to make a positive identification.
In these circumstances, the probative force, if any, of the propensity evidence needed some feature which tended to establish that it was the accused who was this offender. If the propensity evidence possessed such a probative feature then, by hypothesis, it had the capacity to support the prosecution case on the issue of identity. That very feature meant that it also had the capacity to reinforce, supplement, support or 'bolster' (in the sense in which Healy DCJ used that word) the evidence of other prosecution witnesses. However, this tendency to reinforce or bolster the evidence of other witnesses, and so the case for the prosecution, is not its true effect nor is it a reason for admitting that evidence. It is not a basis for telling the jury that it could be acted upon.
Such a description of the effect of the evidence amounts to no more than saying that the evidence itself was 'probative' because, if it were probative, it would have the effect of supporting the other evidence upon which the prosecution relied. However, it is not the relationship between the propensity evidence, and other evidence in the case, which makes the former admissible, but the capacity of the propensity evidence, whether taken alone or in conjunction with other evidence in the case, to prove the fact alleged - in this case the identity of the offender.
None of the other evidence for the prosecution involved establishing some recognisable modus operandi of the offender who stabbed Constable Giocas, nor some particular hallmark, special feature or distinct stamp about the commission of the offence charged. The other witnesses were simply eyewitnesses of what had happened, or in Mr Strachan's case, of recognition and of a private admission from the appellant. They did not reveal any particularly distinctive circumstances about the commission of the offence charged. The events, beginning with the offender's unlawful entry into Mr Yap's car in the carpark, his discovery and pursuit, and the sequence of events culminating in the stabbing are the features which must be examined in order to address the question of similarity, distinctiveness or underlying unity, and therefore probative force, of the propensity evidence which the appellant submits was wrongly admitted.
This 'propensity' evidence shows that the appellant had, on occasions before 3 April 2003 and on an occasion some two weeks later on 16 April 2003, stolen cars and when detected had been chased by police and attempted to escape using the stolen cars. In the process of doing so he had rammed police vehicles, had then attempted to escape on foot, had threatened police officers and members of the public who had come to attempt to apprehend him, including assaulting pursuers with a capsicum pepper spray and making threats of a knife attack and, when further pursued and tackled by police, he had resisted violently and had assaulted a police officer by biting his finger severely. It was also part of the 'propensity' evidence that, on an earlier occasion in the presence of the appellant, his counsel, when making a plea in mitigation to a court dealing with one of those other offences, stated that the accused regularly carried a pocket knife.
The question is whether the propensity evidence had any significant probative effect on the issue of the identity of the appellant as the offender who had attacked and stabbed Constable Giocas.
I do not consider that it did have such a probative effect, and certainly not a significant probative effect. It is true that the evidence did reveal that the accused had habitually stolen motor vehicles in the past, had attempted to make good his escape when pursued, had resorted to violence or threats of violence when attempts were made to overcome him and had in his possession on some of these occasions a small knife. Excluding the attempts to escape by driving away at speed in a stolen vehicle, and ramming police vehicles in an attempt to disable pursuit, the other features described were certainly present in the sequence of events which led to the stabbing of Constable Giocas at the Perth Library plaza. Nevertheless, I do not consider that they have any particular identity, special feature or unique stamp. Sadly, conduct of this kind is experienced in many cases of stealing or attempted stealing whether from a motor vehicle or from some other location or person. Threats or resistance to facilitate flight by an offender are frequently encountered with many crimes. Violence, including assault with a knife, is, regrettably, a frequent feature of many attempts to resist arrest by the police. No doubt the pattern of resistance, or the threats or weapons used, on certain occasions, may have some distinctive stamp or recognisable quality if the offender employs some unusual weapon or some identifiable but uncommon stratagem to effect his attempted escape. For example, an offender who went armed with a 'knuckle duster' or who preyed on a particularly susceptible class of victims, or who committed an offence at the same place or in the same locality repeatedly may perhaps involve sufficient features to render his conduct recognisable, distinctive or otherwise to possess some feature which allows a logical connection to be made which would tend to exclude the likelihood that the offence charged would have been committed by some person other than the person shown to have committed previous offences the subject of the propensity evidence.
Turning to the second phase of the exercise, the balancing between the value and potential prejudicial effect, it is clear that the potential prejudicial effect of this evidence would be very severe. By the admission of the propensity evidence, the appellant was shown to have been a frequent violent offender, who had gone to extreme lengths of recklessness and violence to escape arrest after being discovered stealing a vehicle or vehicles, and who had actually assaulted a police officer and a member of the public when being apprehended for those crimes. Because of s 31A(1)(b), that prejudicial effect would not prevent the admission of the evidence if it was thought that, from the viewpoint of a fair‑minded person, the public interest in receiving that evidence 'of guilt must have priority over the risk of an unfair trial'. In other words, this means that even if the propensity evidence were extremely prejudicial, it could nonetheless be admitted if it amounted to substantial probative evidence of guilt. Or, as Toohey J explained in the passage cited above from Pfennig, the test is not one of whether or not the prejudicial effect outweighs the probative value. As his Honour there explained, and as the statute provides, if the evidence is significantly probative, then, if a fair‑minded person thinks that the public interest of receiving all evidence of guilt justifies its reception, it should and will be received. In other words, disproportion between a severe prejudicial effect and significant probative value will not necessarily lead to the exclusion of the propensity evidence.
In my opinion, one does not get to this second phase in determining the admissibility of this propensity evidence because, for reasons already indicated, I do not consider that it was probative of the fact in issue, let alone significantly probative. However, had I concluded that the evidence was significantly probative, then I consider that this would have been a case where, notwithstanding its prejudicial effect, it should have been admitted, despite the risk to the trial process of its prejudicial effect because of its important potential to reinforce the prosecution case which, apart from the evidence of Strachan, was almost entirely circumstantial. It is in the public interest that evidence which may make the difference between an offender escaping conviction because of weaknesses in other evidence in the case should be received and considered by the jury. But that interest is always dependent upon the evidence which is sought to be admitted being probative.

Accordingly, in my respectful view, the evidence was wrongly received and there is no basis for concluding that it did not cause substantial prejudice to the appellant. It follows that there was a miscarriage of the trial. That is enough to uphold the fourth ground of appeal.
The seventh ground of appeal relates to the manner in which the learned trial judge directed with respect to the propensity evidence. There are a number of passages in the directions to which reference was made - see pages 636, 644, 662 ‑ 663, 670, 674 and 684.
Despite the fact that his Honour was not obliged to do so by the terms of s 31A, the learned trial judge did in fact give a warning about the use and reliance on propensity evidence. This warning encapsulates the rival and inconsistent themes in the direction with respect to propensity evidence of which the appellant complains. The passage is at 684, where the learned trial judge said, in different words but by way of repetition:
Just going back again to Mr Strachan's evidence and the way that the state relies upon the evidence of people who recognised Mr Dair and the telephone calls and the previous incidents, those were led for the purpose of establishing the objective improbability of Mr Dair with the history that you heard about of the other offences and the identification - not the identification but the picking out of photographs of him as being the person involved at the Cultural Centre who is alleged to have done the things that the state alleges he has done as being chosen as the offender when in fact he wasn't the offender, so it's to bolster Mr Strachan's evidence. It's not to be used in any way to look at Mr Dair as a person being of such a character that he is likely to have committed these offences.
The state are not relying upon it for that purpose at all. It's simply to lead the evidence to establish the improbability of the person involved who did those things not being the person Mr Strachan was talking about.
In the first place there are some unfortunate errors of fact in the learned trial judge's direction because none of the witnesses (except for Mr Strachan) of whom he was speaking 'recognised' Mr Dair, nor did they pick out photos of him as being the person involved in the offence but, rather, pointed to photographs indicating that the subject in the photograph was, as best they could tell, similar to the person whom they had seen commit the offence. Those overstatements are no doubt prejudicial to the appellant but, taken in context, it seems likely that the jury were by then fully aware that this particular evidence did not amount to positive identification or recognition and, at the most, only raised circumstantial evidence of identification.
However, the direction in relation to propensity evidence was that this evidence was not to be used in any way to look at Mr Dair as a person being of such a character that he is likely to have committed these offences. Rather, according to the learned trial judge, the evidence could only be used to establish the improbability of the person involved not being the person described by Mr Strachan or, in other words, 'it's to bolster Mr Strachan's evidence'. It seems to me, with respect, that this direction was prone to confuse the jury rather than to convey the true potential significance of the propensity evidence and the limited use which could be made of it.
As observed by McLure JA in Noto v The State of Western Australia [2006] WASCA 278; (2006) 168 A Crim R 457 [24]:
The probative force of admissible propensity evidence, assessed by reference to objective probabilities, derives, directly or indirectly, from propensity reasoning.
Her Honour had already noted that under Australian law propensity evidence is admissible solely to demonstrate a relevant propensity or disposition in exceptional circumstances, citing Pfennig v The Queen [1995] HCA 7; (1995) 182 CLR 461, 481, 484 ‑ 485 and Harriman v The Queen (1989) 169 CLR 590, 598 ‑ 601.
It therefore was the case that the only utility which could be derived from the evidence of Dair's involvement in four other car thefts involving violent resistance to police or persons attempting his apprehension, was that he had a propensity or tendency to act in this way and that this was sufficiently distinctive or apparent to indicate that he had done so on this particular occasion.
The reference to bolstering the evidence of Strachan must be taken as indicating that the propensity just described, taken in conjunction with the evidence of Mr Strachan, makes it more likely that the account given by Strachan is correct. I am satisfied that that must follow, but that does not amount to any more than saying that the propensity evidence taken in conjunction with Strachan's evidence, or for that matter other evidence in the case, renders it likely that it was the accused who committed this particular offence. The evidence 'bolsters' not only Strachan's evidence but the evidence of the other witnesses and the prosecution case generally. That approach is only permissible where the particular probative significance of the propensity evidence has the capacity to do that rather than achieving the result because of the effect of the prejudice which disclosure of prior similar convictions inevitably must have.
This brings me to the direction that the learned trial judge gave with respect to how the propensity evidence could not be used to reason that, because of the character of the accused, it could be concluded that he committed this particular offence.
Where evidence of prior convictions or bad character is received before a jury, in permissible circumstances, care must always be taken to warn the jury clearly that it does not automatically follow that because a person has committed similar offences on previous occasions, he has committed the particular offence with which he is charged. The danger that such a process of reasoning will arise surreptitiously because of prejudice, conscious or unconscious, is so great in the experience of the law that this exclusionary rule is maintained and the warning should be given on every occasion when, exceptionally, such evidence might be improperly used. But that is not to say that it must be given in every case where propensity evidence is admitted.
Propensity evidence involves a process of reasoning that because of the particular propensity or character which has been demonstrated by the prior conduct, it is likely that that offender has committed the particular offence. The distinction between propensity emerging from the manner in which prior offences had been committed or the character of the accused, and mere reasoning by prejudice from the fact of prior convictions can be a very narrow one and, despite the difficulties in identifying and isolating it, it is important that it should be maintained and, when necessary, the jury warned against it. This direction did not have those safeguards. It was confusing because the jurors were told that they could not use the evidence for the very purposes for which propensity evidence is admitted, yet they were also told that the evidence could be used to bolster the evidence of Mr Strachan. That failed to put the position correctly and was likely to cause confusion and to lead to impermissible prejudicial reasoning.
In my respectful view, when propensity evidence in a situation such as the present is admitted, it is necessary for the trial judge to give a careful direction which clearly identifies what, is the alleged probative effect of the particular evidence and which explains to the jury how the evidence might be employed for that purpose if the jury chooses to accept it and to act on it. In doing so, however, the jury also needs to be warned, in terms appropriate to the individual case, that mere bad character or prior convictions could not of themselves justify a conclusion that the accused had committed this particular offence. It should be for the learned trial judge to identify and elicit what is distinctive about this particular propensity evidence which allows a legitimate use to be made of it without allowing the judgment to be influenced by prejudice. That was not done in this case and the absence of that direction left a very high risk of the propensity evidence being misused.
Therefore, even if otherwise admissible (which I have concluded it was not), this propensity evidence was not accompanied by a sufficiently clear direction from the trial judge to inform the jury of just how and the extent to which the evidence could be applied. In my respectful view, the explanation given amounted to a misdirection. Therefore, I would also uphold the seventh ground of appeal and set aside the conviction and order a new trial.
The eighth ground of appeal is that Mr Strachan should not have been permitted to give evidence of his knowledge of the appellant and features and how the appellant may have walked and that he purported to usurp the function of the jury on the question of identification and the question as to whether the surveillance and identikit picture was of the accused. In short, the contention is that Strachan usurped the function of the jury in purporting to make his recognition of the accused conclusive.
Counsel for the appellant relies on Smith v The Queen [2001] HCA 50; (2001) 206 CLR 650. Whether leave to appeal on this ground should be granted or refused was reserved by Wheeler J to this court when sitting on the appeal. I am satisfied that there is no basis to suggest that Strachan's evidence of recognition could be regarded as conclusive and so usurping the function of the jury. This present case is entirely distinguishable from Smith v The Queen and the question of whether or not it was the appellant on the video and whether Strachan's evidence was correct in contending that it was, remained in contention throughout the trial and was obviously a matter for decision and resolution by the jury in its deliberations.
I see no prosects of success for this ground and I would refuse leave. If otherwise leave were granted I would reject this ground of appeal.

Appeal against sentence

Miller JA has set out comprehensively the sentence imposed on Dair by the learned trial judge, the reasons for that sentence and the appellant's lengthy previous record of serious offending and it is unnecessary for me to repeat any of those details.
In view of the conclusion which I have reached that the appeal against conviction should be allowed, I must address the appeal against sentence on the hypothesis that the conviction stands. On that approach, I can only say that, with respect, I agree with the reasons given by Miller JA for concluding that the appeal against sentence should be allowed and with his Honour's conclusion that, having regard to the principle of totality, a shorter sentence should be substituted. On this approach, therefore, I agree with the orders proposed by Miller JA for the imposition of a shorter sentence should that be necessary.