SUPREME COURT OF NEW ZEALAND
WILLIAM YOUNG J
19 MAY 2011
(with whom Elias CJ and Blanchard J)
This appeal is concerned with what the Evidence Act 2006 calls propensity evidence. In his reasons William Young J has comprehensively set out the circumstances of the case and the issues that arise. We can therefore move straight to a consideration of those issues.
Section 40(1) defines propensity evidence as evidence that tends to show a person’s propensity to act in a particular way or to have a particular state of mind, being evidence of acts, omissions, events or circumstances with which a person is alleged to have been involved. Propensity evidence does not include evidence of an act or omission that is one of the elements of the offence for which a person is being tried. The definition of propensity evidence covers a propensity in relation to a specific person as well as a propensity to behave or think in a particular way as regards persons more generally.
The rationale for the admission of propensity evidence rests largely, as William Young J says, on the concepts of linkage and coincidence. The greater the linkage or coincidence provided by the propensity evidence, the greater the probative value that evidence is likely to have. It is important to note, however, that the definition of propensity evidence refers to a tendency to act in a particular way or to have a particular state of mind. It is necessary, therefore, that the propensity have some specificity about it. That specificity, in order to be probative, must be able to be linked in some way with the conduct or mental state alleged to constitute the offence for which the person is being tried.
We do not consider a great deal is now to be gained from an examination of pre-Evidence Act case law. The Act substantially codified that case law and it is preferable, and consistent with s 10(1), to focus firmly on the terms of the Act; albeit the application or interpretation of a particular provision in the Act may sometimes benefit from a consideration of the previous common law.
Section 43 deals with cases like the present where propensity evidence is offered by the prosecution about defendants. The prosecution may do this only if the evidence has a probative value, in relation to an issue in dispute, which outweighs the risk that the evidence may have an unfairly prejudicial effect on the defendant. In s 43 cases the prosecution must satisfy the court that the criterion for admissibility is demonstrated. If that is shown there cannot be any basis for exclusion under s 8 which deals in general and materially similar terms with a similar issue.
Section 43(3) sets out a number of matters the court may consider in assessing probative value. Most, if not all, of these matters were conventionally taken into account at common law. Clearly the subs (3) matters are not exclusive (“among other matters”). Although these matters are not couched as mandatory considerations it will usually be appropriate to consider them, to the extent they are relevant, in the course of making the necessary assessment. Some of them are factors referable to conduct with a person or persons other than the complainant, and may have no application when the conduct in issue involves only the complainant.
In order to make the necessary assessment the court must carefully identify how and to what extent the propensity evidence has sufficient particularity to be probative, and how and to what extent it risks being unfairly prejudicial. Obviously any evidence that is probative will be prejudicial to the accused but not normally unfairly so. Unfairness is generally found when and to the extent the evidence carries with it a risk that the jury will use it for an improper purpose or in support of an impermissible process of reasoning. In assessing the probative value/unfair prejudice balance, the court may need to take into account the extent to which it considers a “proper use” direction in the trial judge’s summing-up is likely to guard against the risk of improper use. With those introductory remarks, we turn to the circumstances of the present case.
The van incident was evidence tending to show that Mr Mahomed, and to a lesser extent Mrs Mahomed, had a particular state of mind. That state of mind was one of lack of care or foresight as regards the welfare of their then ten-week-old child. It was a state of mind which demonstrated indifference, perhaps recklessness, as regards the child’s welfare but it would be a long stretch to suggest that this evidence was capable of demonstrating active hostility towards the child leading to violence against her. As we indicated earlier, it is necessary to identify with some specificity the “particular” state of mind the propensity evidence tends to show and relate that to the states of mind required for each offence.
This case is complicated by the fact that Mr Mahomed faced two counts of causing grievous bodily harm to the child (counts one and two), one count of murder of the child (count three) and one count (jointly with Mrs Mahomed) of failing to provide the necessaries of life for the child (count four). It does not follow that the probative value/unfair prejudice assessment will necessarily yield the same result in respect of all counts. We do not consider the van incident could reasonably be said to be probative as regards counts one and two (grievous bodily harm), save in respect of the limited aspect of Mr Mahomed leaving the scene in a hurry, inferentially in order to avoid an encounter with the police or some other person in authority. But the probative force of the evidence in that respect was comparatively slight and it would be difficult to set the scene adequately for the abrupt departure without giving contextual evidence of the substance of the van incident which did not, in our view, satisfy the s 43(1) criterion for admissibility.
Overall we consider the particular state of mind demonstrated by the van incident was not sufficiently probative of the state of mind necessary to establish counts of intentionally causing grievous bodily harm (counts one and two) so as to outweigh the risk of unfair prejudice inherent in offering evidence of the van incident. That risk was of the jury, despite any directions given by the Judge, being improperly influenced against the appellants because the van incident showed them as uncaring parents. The level of probative force in the van incident as regards counts one and two was such that relatively little unfair prejudice was capable of outweighing it.
By way of contrast, we consider the evidence of the van incident was sufficiently probative, as against unfairly prejudicial, to be admissible against both Mr and Mrs Mahomed in respect of the joint charge against them of failing to provide the necessaries of life. The mental states in each have distinct similarity. What the jury made of the evidence was for it. But in respect of that charge, and this is the only charge which Mrs Mahomed was facing, the evidence was admissible.
The murder charge (count three), of which Mr Mahomed was convicted, is obviously the most serious. We consider the evidence of the van incident was not probative as regards the identity, as between the parents, of the person who inflicted the fatal injuries; nor was it sufficiently probative of the state of mind necessary to make that person a murderer, whether on account of an intent to kill or an intent to cause bodily injury with the necessary degree of recklessness as to death ensuing. There is really no, or very little, probative coincidence or linkage between thoughtlessly leaving a child in a van and causing the child’s death with murderous intent. It may be said that Mr Mahomed’s conduct at the end of the van incident was capable of being circumstantial evidence indicative of a guilty state of mind but that can only have been in relation to events which preceded the van incident. If the van incident was probative at all in those respects, the evidence did not, in our view, satisfy the need for its probative value to outweigh the risk it may have had of causing unfair prejudice to Mr Mahomed.
In making these assessments we have considered the s 43(3) matters. There is no frequency factor (para (a)); nor is there any significant connection in time (para (b)). The mental state demonstrated by the van incident has little, if any, similarity to the mental states necessary for the offences charged in counts one, two and three (para (c)). There is no question of multiple complainants (para (d)); nor can there be any suggestion of collusion or suggestibility (para (e)). The final factor of unusualness (para (f)) does arise because it is unusual, in general terms, for parents to be negligent or lacking consideration for the welfare of their very young child. But in the total sum of things this is not a persuasive factor in favour of admissibility on the charges involving specific intent.
So we have reached the point where the evidence was admissible only as regards count four. It is necessary therefore to examine carefully how the trial Judge directed the jury as to the use it could make of the van incident evidence. We do not consider the Judge’s directions gave the jury the necessary assistance in that respect. Understandably he did not direct the jury that the van incident was relevant only to count four. That would have been contrary to the pre-trial ruling of the Court of Appeal which was to the effect that the van incident was admissible generally: R v Mahomed  NZCA 477. As regards the count of murder, the Judge directed the jury as follows:
The fourth area is the car park incident. Mr Hamlin says it is directly relevant to the murder charge. That is for you to determine. The issue is whether you consider it shows that Mr Mahomed was in the habit of acting negligently towards Tahani and had an uncaring attitude towards her. Whether or not you reach that conclusion or whether you form a view that in all the circumstances he was prepared to risk the baby’s health, even her life, by leaving her unattended in a hot car while he worked is for you to decide. What you make of his apparent indifference to Ms Trevena’s reference to Tahani’s condition when he arrived is again for you to decide. Mr Hamlin says this incident illustrates how Tahani had assumed a nuisance value for her parents. Again whether you agree or whether you think it is relevant to the murder charge is entirely for you.
These directions would, with respect, have been problematical even if the van incident had been admissible generally. The Judge told the jury that the issue was whether the evidence showed that Mr Mahomed was in the habit of acting “negligently” towards the child and had an “uncaring attitude” towards her. He then referred to “apparent indifference” and “nuisance value”. The Judge left it to the jury to decide whether the evidence was relevant to the murder charge. He did so without giving them any guidance as to what would be an appropriate basis for that conclusion and what would not. The references to negligence, uncaring attitude and indifference were potentially misleading without an express reminder that these states of mind were not in themselves sufficient to sustain a murder charge or one involving intent to cause grievous bodily harm, coupled with a reminder of what the necessary states of mind were and an explanation of how the lesser states of mind referred to might be relevant to the existence of the necessary state of mind. The latter would not have been an easy task.
We are satisfied that a miscarriage of justice occurred, not from the admission of the evidence, because it was admissible in respect of count 4, but from the Judge’s understandable failure to identify that it was admissible only on that count. This was compounded by the difficulties in the passage from the summing-up to which we have just referred.
We would prefer to say no more than we have on the general question of jury directions in propensity cases addressed by William Young J in his reasons. It is preferable to deal with particular problems and issues as and when they arise.
The question becomes whether the Court should apply the proviso to s 385(1) of the Crimes Act 1961. In terms of R v Matenga  NZSC 18,  3 NZLR 145 the Court cannot apply the proviso unless satisfied the verdicts of guilty on counts one, two and three were inevitable in the sense of being the only reasonably possible verdicts on all the admissible evidence. Hence we must feel sure of the guilt of Mr Mahomed on those counts. The case was not said by counsel to raise fair trial issues in the sense referred to in Matenga.
We start with the murder count. The injuries to the child were such that the only reasonably possible inference is that they must have been inflicted with murderous intent. If they were not inflicted with an intent to kill, they most certainly must have been inflicted with recklessness as to whether death ensued or not. The only other issue is who inflicted those injuries. Does the evidence show beyond reasonable doubt that it was Mr Mahomed rather than Mrs Mahomed who did so? We consider it does. The injuries which were the subject of counts one and two were also such that it is an irresistible inference that whoever inflicted them must have done so with intent to cause grievous bodily harm. With these counts also we are satisfied that the evidence demonstrates beyond reasonable doubt that it was Mr Mahomed who inflicted these injuries rather than Mrs Mahomed. The van incident evidence was, of course, admissible on count four and there is therefore no basis for finding a miscarriage of justice on account of its admission on that count. The Judge’s inadequate directions can have had no effect on the jury’s reasoning on that count, and so no miscarriage of justice can have arisen. Hence no question of applying the proviso arises in that respect.
The consequence is that both Mr Mahomed’s appeal and that of Mrs Mahomed must be dismissed.
McGrath J & William Young J
Following trial before Harrison J and a jury, Mr Azees Mahomed was found guilty of the murder of his 11-week-old daughter Tahani (count three) and of twice intentionally causing her grievous bodily harm (counts one and two). Mr Mahomed and Tahani’s mother, Mrs Tabbasum Mahomed, were found guilty of a further charge (count four) of failing to provide Tahani with the necessaries of life (namely prompt medical treatment). Their appeals against conviction and sentence were dismissed by the Court of Appeal (Mahomed v R  NZCA 419 [Mahomed – Conviction appeal]) and they now appeal to this Court.
Their appeals turn on:
the admissibility of evidence that eight days before Tahani was fatally assaulted, Mr and Mrs Mahomed left her in a van at the Otahuhu shopping centre where she became distressed and as to their subsequent behaviour; and
the way in which this evidence was dealt with by the trial Judge.
The core facts
Mr and Mrs Mahomed came to New Zealand from South Africa in 2006. They were accompanied by their daughter Tasmia who was then one year old.
Tahani was born on 7 October 2007. She was seen by a Plunket nurse on 15 November 2007 when she received a vaccination. She otherwise had no contact with the health system until her parents brought her into Middlemore Hospital at 8.11 am on 28 December 2007. She was very ill and, as well, was distinctly malnourished.
The medical evidence showed that Tahani had suffered some very serious injuries:
Injuries to her brain, with associated likely blindness in her left eye and retinal damage to the right eye likely to have been inflicted around 2–5 weeks before admission.
A fractured shin, suffered around 1–2 weeks before admission.
Recent injuries (probably inflicted around 12 hours before admission) consisting of a fracture to the skull at the back of her head, associated brain damage and a fractured thigh bone.
Tahani died on 1 January 2008 as a direct consequence of the third set of injuries.
Police officers interviewed Mr and Mrs Mahomed separately on 28 December 2007. At interview, Mr and Mrs Mahomed gave very similar accounts: Tahani was fed normally between 8.00 pm and 9.00 pm on 27 December, she was “startled” or “frightened” before going to sleep, she did not want to wake up to be fed at midnight or at 3.00 am and would not take milk at 5.00 am. Mrs Mahomed then rang her doctor and Healthline seeking advice and was told to take her to hospital. This they did. Apart from Tahani not wishing to wake up and not taking milk, she was well when they arrived at hospital. Both described Tahani’s life as having been uneventful and her health as being good.
The grievous bodily harm charges focussed on the first and second of the injuries listed in . On the Crown case these injuries and the ultimately fatal injuries were inflicted by Mr Mahomed. The charge of failing to supply the necessaries of life related to the last 12 hours of Tahani’s life and the failure by Mr and Mrs Mahomed to seek prompt medical treatment. On the Crown case, the reason for this failure was a concern on their part that seeking medical treatment would result in Tahani’s injuries and perhaps her malnourished state coming to official notice.
Although count four encompassed both 27 and 28 December 2007, there was considerable focus on the period between 6.00 am and 8.00 am on 28 December. The evidence showed that Mrs Mahomed telephoned her general practitioner and the Healthline Service just before and just after 6.00 am on 28 December but despite both the doctor and the Healthline operator saying that Tahani should be taken to hospital straight away, Mr and Mrs Mahomed did not take her there for another two hours. The drive from their residence to the hospital would have taken only seven or eight minutes.
The intercept evidence
From the outset, it was very plain that Tahani had been seriously assaulted. She had only ever been in the care of Mr and Mrs Mahomed and her injuries could only have been inflicted by one or other or both of them. The accounts of events which they had given on the night of 28 December 2007 indicated that at that point they were acting collusively; this given the consistency of each account with the other but the marked inconsistency of those accounts with Tahani’s condition when she arrived at hospital. What was not clear, however, was who had inflicted Tahani’s injuries.
The Crown contention that it was Mr Mahomed who had inflicted Tahani’s injuries was primarily based on face-to-face conversations involving Mr and Mrs Mahomed between 3 and 30 January 2008. These discussions were intercepted pursuant to a warrant obtained by the police. One of the intercepted discussions makes it clear that Mr and Mr Mahomed were conscious of the risk that their phone calls were being intercepted. Although there is no indication in the intercepts that they also appreciated that there might be interception devices in their house, Mr and Mrs Mahomed sometimes kept their voices low and for this and other reasons (associated with the particular location of the devices) there are gaps in what was recorded. And by way of further difficulty, while some of the discussions were in English, most were in Hindi (but with some Urdu and Gujarati words) and thus required translation for the jury. There was some dispute at trial about the translation (see Mahomed – Conviction appeal at –).
Although there are ambiguities and uncertainties in relation to the intercepts, we are satisfied – and indeed counsel for the appellants accepted – that it was open to the jury to conclude that:
it was common ground between Mr and Mrs Mahomed that Mr Mahomed had inflicted the critical injuries;
they acknowledged that they had not properly looked after Tahani; and
they proposed concocting innocent explanations for what had happened to Tahani and coaching Tasmia so that if she were interviewed, she would not implicate Mr Mahomed.
The van incident
On the morning of 19 December 2007 Mr and Mrs Mahomed were at the Otahuhu shopping centre selling jewellery. While they were doing so, they left Tahani in their van. This was despite it being a warm day. On the basis of the evidence as a whole it was open to the jury to conclude that she was left in the van at about 9.00 am. At around midday, Ms Janice Trevena, the shopping centre’s security officer, came to the van. The front passenger window was down slightly but the others were up. Tahani was crying and her face was covered with perspiration. Ms Trevena called 111 and while she was on the phone Mr Mahomed came over. He did not indicate any concern about Tahani’s condition. She and Mr Mahomed had an argument as to whether what was on Tahani’s face was perspiration (as opposed to spilt milk as Mr Mahomed claimed). Ms Trevena tried to keep Mr Mahomed there until the police arrived but he insisted on leaving.
Although Mrs Mahomed was not with Mr Mahomed when he returned to the van, she obviously learnt of the incident shortly afterwards. This is because, not long after midday, she told a Mr Walter Turner that she was in trouble and that the police were coming. She also mentioned something about her baby being in the car.
On the Crown case Tahani was, by 19 December 2007, already carrying the injuries referred to in (a) and, in all probability, the second injury (ie that referred to in (b)) as well. Had the police (and consequently social welfare officials and doctors) become involved with Tahani on 19 December, her injuries and perhaps her malnourished state would probably have been discovered. Concern that this might happen provides a plausible explanation for Mr Mahomed driving off without waiting for the police to arrive.
The van incident was referred to several times in the intercepted communications and it is clear that Mr Mahomed at least regarded the incident as amounting to “one very big proof against us”.
The procedural history
Mr and Mrs Mahomed originally faced a charge of failing to provide the necessaries of life in relation to the van incident and a similar charge in relation to Tahani’s general lack of nourishment. In a pre-trial ruling, however, Chisholm J directed that these counts be tried before a judge alone (which in substance involved severance of those counts from the four counts on which Mr and Mrs Mahomed were ultimately tried before Harrison J and a jury): R v Mahomed HC Auckland CRI-2008-092-748, 16 July 2009 at . He also ruled that evidence as to the van incident was not to be led before the jury, but that the Crown could lead evidence as to Tahani’s malnourished state.
The Crown had no right of appeal in relation to the direction that the two counts in question be tried by judge alone but appealed successfully against the evidential ruling: R v Mahomed  NZCA 477 [Mahomed – Admissibility appeal].
In the course of its judgment allowing the Crown appeal, the Court of Appeal discussed the admissibility of the evidence in terms of the propensity rule under ss 40–43 of the Evidence Act 2006: Mahomed – Admissibility appeal at –. It noted that the van incident could be seen as showing a propensity on the part of Mr Mahomed to act towards Tahani “in a way that was careless of her wellbeing and indifferent to her needs and suffering, and also indifferent to the consequences of such behaviour towards her”. It also accepted, in the same paragraph, that the evidence as to this incident could be said to show a propensity on the part of Mrs Mahomed “to go along with ill-treatment of Tahani by Mr Mahomed” (at ). The Court, however, did not carry its propensity analysis through to a conclusion because it considered that there was no point in doing so as:
[It] would overlook the most relevant aspect of the behaviour, which is its direct relevance .... The central probity is not in showing propensity in the sense of a tendency to carry out particular acts that were similar; but rather to show a train of neglect leading up to Tahani’s death. While the evidence can be seen as probative as showing propensity, its probity is best weighed as part of the relevant facts.
The direct relevance identified by the Court of Appeal (at –) was as follows:
The incident occurred in the middle of the maltreatment which the Crown alleged had led to the fatal injuries suffered by Tahani and tended to show that Mr Mahomed was engaged in a neglectful course of conduct towards Tahani as well as being indifferent to her wellbeing. His hasty departure from the scene could be seen as associated with a concern that a check on Tahani would have revealed the injuries she was already carrying. It was also indicative of an attitude towards Tahani which was consistent with what the Crown alleged in relation to the murder count, in that it showed a willingness to harm her and leave her injured as well as being generally indifferent to her welfare.
In the case of Mrs Mahomed, the van incident indicated reasonably quick communication between Mr and Mrs Mahomed in relation to conduct which on the Crown theory involved abuse of Tahani and that in such circumstances, Mrs Mahomed was more concerned about police involvement than Tahani’s welfare. As well, given that on the Crown case Tahani was already carrying serious injuries, the behaviour of Mrs Mahomed was indicative of an attitude which was consistent with the Crown case against her on the count of not providing the necessaries of life associated with the events of 27–28 December. Further, the evidence was also inconsistent with at least the drift of what Mrs Mahomed had told the police about what she had claimed was the uneventful nature of Tahani’s life.
In ruling that the evidence was admissible the Court also addressed s 8(1)(a) of the Evidence Act and in particular considered whether admitting the evidence would have an unfairly prejudicial effect on the proceedings. In carrying out this exercise, the Court acted by way of analogy with the s 43(4) test which applies to propensity evidence and concluded that the evidence was admissible: See –.
As a result of the ruling that the two counts of failing to supply the necessaries of life associated with the van incident and Tahani’s malnourished state were to be tried by judge alone, the indictment on which Mr and Mrs Mahomed stood trial before Harrison J and a jury contained only the four counts on which they were ultimately found guilty. But, on the basis of the Court of Appeal’s admissibility ruling, the Crown was able to lead the evidence as to both malnutrition and the van incident in relation to those counts. So the severing off of the associated failing to supply the necessaries of life counts made no difference to the evidence which the Crown led.
A primary plank in the defence advanced on behalf of Mr Mahomed was that the person who inflicted the injuries was, or may have been, Mrs Mahomed. As is apparent, Mrs Mahomed did not face a murder charge. But if she had inflicted the fatal injuries, her delay in taking Tahani to hospital might be thought necessarily to have involved a criminal failure to supply the necessaries of life. Despite this, there was no attempt on her part at trial to push back against the contention that she had inflicted all the injuries to Tahani. Instead, there was a joint and co-ordinated defence strategy.
In his closing address to the jury, the prosecutor Mr Hamlin referred to and relied on the van incident in two broad ways:
He suggested that the driving off by Mr Mahomed (in particular his refusal to wait for the police) and the anxiety on the part of Mrs Mahomed were a result of concern that if the police had become involved, Tahani’s then injuries would have become apparent. In this respect, he claimed that Mr Mahomed’s actions in driving off reflected a guilty conscience.
Generally, and to some extent relying on what was said about this incident in the intercepted communications, he treated this incident, Tahani’s malnourishment and the failure to secure prompt medical attention for Tahani on 27 and 28 December as part of a pattern of events showing consistent behaviour on the part of Mr and Mrs Mahomed towards Tahani.
In his closing address, Mr Borich, for Mrs Mahomed, conceded that she had not been a good parent but challenged the Crown’s factual allegations as to the detail of the van incident and contended that the incident was not material to the charge faced by Mrs Mahomed at trial of failing to supply the necessaries of life.
Mr Wilkinson-Smith for Mr Mahomed also conceded that Mr and Mrs Mahomed had not been good parents. He maintained that the van incident did not have the tendency to show which of Mr and Mrs Mahomed had inflicted the fatal injuries. He adopted the arguments advanced by Mr Borich in relation to the charge of failing to supply the necessaries of life.
In his summing-up, the Judge gave standard directions in relation to emotion and prejudice but specifically addressed the van incident only in passing. The primary discussion was in these terms:
.... The issue is whether you consider it shows that Mr Mahomed was in the habit of acting negligently towards Tahani and had an uncaring attitude towards her. Whether or not you reach that conclusion or whether you form a view that in all the circumstances he was prepared to risk the baby’s health, even her life, by leaving her unattended in a hot car while he worked is for you to decide. What you make of his apparent indifference to Ms Trevena’s reference to Tahani’s condition when he arrived is again for you to decide. Mr Hamlin said this incident illustrates how Tahani had assumed a nuisance value for her parents. Again whether you agree or whether you think it is relevant to the murder charge is entirely for you.
A little later, commenting on Mr Wilkinson-Smith’s contention that the van incident reflected badly on Mr Mahomed but otherwise was of no moment, the Judge said:
.... But he says Mr Mahomed is not on trial for being a bad parent. He says the question is whether or not they committed these offences. He says that what happened in the car park is not directly relevant to consideration of whether or not Mr Mahomed inflicted the fatal blow on his daughter.
When dealing with the charge of failing to supply the necessaries of life, the Judge mentioned the van incident only in a summary way and again by reference to the closing addresses of counsel:
Mr Borich led the defence arguments on this count. He addressed you with great care. He took you through the evidence under eight separate headings. They included .... the Otahuhu car park incident and Tahani’s general state of malnourishment .... Mr Wilkinson-Smith effectively adopted all of those arguments, although he of course would substitute Mr Mahomed’s account to the police for Mrs Mahomed’s.
After the summing-up, defence counsel invited the Judge to give further direction on the van incident in terms of propensity and in particular to do this along the lines approved by the Court of Appeal in Stewart (Peter) v R  NZCA 429,  1 NZLR 197, discussed below at . This he declined to do and he explained why in these terms:
I declined to give a supplementary direction on that point. Mr Hamlin did not address the jury on the footing that the car park incident fell into the propensity category. The relevance of the evidence is set out in the Court of Appeal’s decision in R v Mahomed  NZCA 477 [Mahomed – Admissibility appeal] at –. My directions summarise those passages.
The Court in Mahomed considered the propensity issue on an alternative basis: at –. However, it did not decide the question but instead reaffirmed the true relevance of the evidence, not in its propensity context but as bearing upon the Mahomeds’ treatment of Tahani.
With respect, the evidence of the car incident is not propensity evidence. It is not offered to show Mr Mahomed’s propensity to act in a particular way or to have a particular state of mind or pattern of behaviour (the evidence of his treatment of the older child, Tasmia, is to the contrary), but as discrete evidence of his attitude of neglect of and disinterest in Tahani’s general welfare. A specific direction is unnecessary, and on the facts would serve to confuse rather than enlighten a jury. I am satisfied that the existing direction satisfies the needs of this case.
The judgment of the Court of Appeal on the conviction appeal
Amongst the grounds of appeal advanced on behalf of Mrs Mahomed were contentions that the evidence as to the van incident and Tahani’s malnourishment
was propensity evidence,
should have been addressed as such and
should have been held to be inadmissible.
As well, and in the alternative, it was contended that the directions of the Judge were inadequate.
In dismissing these arguments, the Court of Appeal referred to the procedural history and, in particular, the earlier admissibility decision of the Court of Appeal. In concluding that the evidence was admissible, the Court of Appeal adopted the approach taken in the admissibility appeal. It concluded as well that an elaborate propensity direction along the lines envisaged by Stewart “would not sit comfortably with [the] evidence or the way in which it was used” and saw this as suggesting that “the evidence did not carry with it the dangers that can be associated with propensity evidence”: Mahomed – Conviction appeal at . The Court held (at ) that the directions given by the Judge were sufficient.
Was the evidence about the van incident subject to ss 40–43 of the Evidence Act 2006?
Evidence of bad character – common law admissibility rules
The common law was succinctly stated by Lord Herschell LC in Makin v Attorney-General for New South Wales  AC 57 (PC) at 65:
It is undoubtedly not competent for the prosecution to adduce evidence tending to shew that the accused has been guilty of criminal acts other than those covered by the indictment, for the purpose of leading to the conclusion that the accused is a person likely from his criminal conduct or character to have committed the offence for which he is being tried. On the other hand, the mere fact that the evidence adduced tends to shew the commission of other crimes does not render it inadmissible if it be relevant to an issue before the jury, and it may be so relevant if it bears upon the question whether the acts alleged to constitute the crime charged in the indictment were designed or accidental, or to rebut a defence which would otherwise be open to the accused.
Many of the cases which followed Makin involved what may be termed orthodox similar fact evidence, usually to the effect that the defendant had, on other occasions, acted in a way which was similar to that involved in the offence for which the defendant was being tried. A typical Crown contention in a case involving alleged sexual offending might be that the defendant was either guilty or the victim of an implausible coincidence (namely to be falsely accused in such a similar way by a number of complainants). Another typical Crown contention in such a case might be that the defence case rested on the implausible coincidence that of all the people the complainant chose to make a false complaint about, someone who just happened to have a proclivity to act in the way alleged was picked. As we will indicate later, this particular propensity theme has many variations, but common to them all are ideas about coincidence and probabilities.
In other cases, evidence of what might be considered bad character was led for reasons which had nothing to do with the propensity of the defendant to act (or think) in a particular way. For instance, where the motive alleged for a murder was that the defendant wished to avoid prosecution for what may have been earlier, distinct, offending against the victim, the background evidence that complaints of such offending had been made and that this was known to the defendant was admissible for that reason and not on similar fact principles. If the alleged offending had occurred in a prison and involved serving prisoners, the defendant’s status as a serving prisoner would necessarily be before the Court.
Rather more difficult to classify were cases where the prosecution alleged other offending by the defendant against the victim.
In this context, the well-known case of R v Ball  AC 47 (HL), although not directly on point, is of interest and makes a useful starting point. The defendants were brother and sister and were charged with incest alleged to have been committed in July and September 1910. The primary evidence against them was circumstantial and perhaps equivocal but there was ample evidence to show that they had engaged in sexual intercourse in the period just prior to criminalisation (via the Punishment of Incest Act 1908) of such conduct as incest. The House of Lords, reversing the Court of Criminal Appeal, held that the evidence of earlier sexual intercourse was admissible. The basis upon which this ruling proceeds is, however, unclear. This is what Lord Loreburn LC said (at 71):
My Lords, the law on this subject is stated in the judgment of Lord Chancellor Herschell in Makin v Attorney-General for New South Wales; it is well known and I need not repeat it — the question is only of applying it. In accordance with the law laid down in that case, and which is daily applied in the Divorce Court, I consider that this evidence was clearly admissible on the issue that this crime was committed — not to prove the mens rea, as Darling J. considered, but to establish the guilty relations between the parties and the existence of a sexual passion between them as elements in proving that they had illicit connection in fact on or between the dates charged. Their passion for each other was as much evidence as was their presence together in bed of the fact that when there they had guilty relations with each other.
My Lords, I agree that Courts ought to be very careful to preserve the time-honoured law of England, that you cannot convict a man of one crime by proving that he had committed some other crime; that, and all other safeguards of our criminal law, will be jealously guarded; but here I think the evidence went directly to prove the actual crime for which these parties were indicted.
This case is usually seen as turning on similar fact principles, particularly in light of the reference to Makin and the need to apply it. But when Lord Loreburn said that “the evidence went directly to prove the actual crime”, he might be thought to have adopted an approach which was broadly similar to that of the Court of Appeal in the admissibility judgment in this case. This view of the case is consistent with remarks made in the course of argument by Lord Atkinson (at 68):
Surely in an ordinary prosecution for murder you can prove previous acts or words of the accused to shew he entertained feelings of enmity towards the deceased, and that is evidence not merely of the malicious mind with which he killed the deceased, but of the fact that he killed him .... Evidence of motive necessarily goes to prove the fact of the homicide by the accused, as well as his “malice aforethought”, inasmuch as it is more probable that men are killed by those who have some motive for killing them than by those who have not.
There are many reasons why the prosecution might wish to lead evidence of misconduct by the defendant towards the victim. In some cases, the misconduct might be inextricably linked to the alleged offending. To take an obvious and uncontroversial example of the defendant having threatened to kill the victim immediately before inflicting the fatal injury, the threat to kill could be regarded (at least by a pedant) as misconduct which betrayed a tendency to think about the victim in a particular way. But the obvious admissibility of such evidence has never been doubted. Other misconduct by the defendant towards the victim, even though not so closely linked in time with the offence, might reveal hostility by the defendant to the victim and thus be evidence of motive. Sometimes, such evidence might have important explanatory value, for instance as enabling a victim to explain why he or she had acted in a particular way. So in cases of alleged intra-familial sexual abuse, a victim may be unable to give a coherent account of what happened and why without discussing the underlying family dynamics.
Evidence of the kinds just discussed differ from orthodox similar fact evidence in a number of overlapping respects:
With orthodox similar fact evidence, the only link between the evidence and the offending in issue is through the particular propensity which the Crown attributes to the defendant. In that respect, the other misconduct is extraneous to the alleged offending. In contradistinction, evidence of misconduct by the defendant towards the victim can usually be seen as having a more direct relevance to the issues which the jury must decide.
The relevance of the evidence of misconduct by the defendant to the victim will normally not depend (at least primarily) on ideas about coincidence and will usually be sufficiently obvious as to not require particular explanation.
A high level of similarity between the alleged offending and the other conduct is not fundamental to probative value. Thus where the allegation is that the defendant murdered the deceased with a firearm, evidence of a prior attempt by the defendant to murder the deceased with a knife is likely to be relevant.
The risk of unfair prejudice to the defendant arising out of evidence of other misconduct to the victim is likely to be less than with orthodox similar fact evidence. This is because the misconduct is usually not extraneous to the alleged offending and thus the associated evidence will not portray the defendant as being generally of bad character. In criminal trials it is routine for evidence of direct relevance to the alleged offending to reflect badly on the defendant, perhaps in terms of the defendant being a gang member, a drug user, an associate of criminals or having anti-social attitudes. The risk of unfair prejudice associated with such evidence is usually addressed simply by the judge warning the jury in general terms against being influenced by prejudice or emotion.
Prior to the enactment of the Evidence Act 2006, New Zealand judicial practice was mixed as to the applicability of similar fact rules to evidence of other misconduct by the defendant to the victim. On occasion, such evidence was dealt with in accordance with similar fact principles without any discussion as to whether an alternative approach might be taken (as in R v Rongonui  2 NZLR 385 (CA)). On the other hand, there are cases where the courts either rejected the applicability of the similar fact calculus in such circumstances (T (CA 175/97) v Attorney-General CA175/97, 27 August 1997) and proceeded on the basis of direct relevance (see for instance R v Meynell  1 NZLR 507 (CA) at ), or the res gestae principle (R v Karetai (1988) 3 CRNZ 564 (CA) and R v Baker  1 NZLR 738 (CA)). As well, there were cases in the middle where it was explicitly recognised that evidence of conduct between the defendant and victim might be admissible on the basis of both similar fact principles and for direct relevance. Further, there was the practical dimension that when a defendant faced a number of charges involving the same victim, those charges would customarily be tried together and it would not normally occur to the participants in such a trial that similar fact principles might apply to the cross-admissibility of the evidence associated with the different charges or that there was any need for any particular propensity directions.
Sections 40–43 of the Evidence Act 2006
Section 40 of the Evidence Act 2006 now provides:
In the case of propensity evidence led by the prosecution against a defendant, the relevant section is s 43.
Some preliminary points. These sections do not purport to cover all the ground as to the admissibility of what the pre-Evidence Act common law termed bad character evidence. For instance, they would not be engaged by the examples given in . As well, it is important to recognise that the propensity definition operates solely to define the type of evidence which the specific admissibility rules provided for in ss 41–43 cover. The definition is not itself an admissibility rule. And associated with this, the definition works on the basis that evidence is “propensity evidence” if it has the tendency which is specified. It is not confined to evidence which is to be led for any particular purpose.
The definition of “propensity evidence” is exhaustive and there is no exclusion in relation to evidence of conduct between the defendant and victim. This means that evidence which tends to show a propensity by the defendant to act towards or think about the victim in a particular way is necessarily within the definition of “propensity evidence” irrespective of why the Crown wishes to lead the evidence. Further, given the direction in s 40(3), evidence which is within the statutory definition can only be adduced if admissible under s 43. So there is no scope for addressing the admissibility of such evidence on the basis of both the propensity rules and direct relevance.
As will become apparent, we do not see any practical difficulties with this approach. Although the s 43 criteria (to which we are about to turn) were plainly drafted with more orthodox (that is traditional similar fact) propensity evidence in mind, they are not exhaustive. Thus the types of relevance invoked in the cases referred to in  can also be invoked under s 43 when assessing and weighing the probative value of evidence against the risk of unfair prejudicial effect on the defendant.
Comparing the s 43(1) test to the s 8(1) test
A major plank in the admissibility arguments of counsel for Mr and Mrs Mahomed was the contention that the test for the admissibility of propensity evidence provided by s 43(1) is more exacting than the general test provided by s 8.
Section 43 of the Evidence Act provides:
Propensity evidence offered by prosecution about defendants
The section 8 test is expressed in these terms:
In s 43 the explicit focus is on the potential for prejudice to the defendant. In contradistinction, s 8 is addressed to “prejudicial effect on the proceeding”. Although this point was relied on by counsel we do not place significance on it. Section 8(1) deals with the general admissibility of all evidence, that is, irrespective of whether the proceedings are civil or criminal and who proposes to call the evidence. On the other hand s 43(1) addresses only criminal proceedings and evidence called by the Crown about a defendant. In the case of such evidence, the only relevant prejudicial effect which logically needs to be considered is the likely impact on the fairness of the trial from the point of view of the defendant. Possibly more significant is the consideration that s 8(1) requires exclusion where the probative value of the evidence is outweighed by the risk of prejudicial effect, whereas s 43 permits propensity evidence only where the probative value of the evidence outweighs the risk of prejudicial effect. So if probative value and the risk of unfair prejudice were equal, exclusion would only be required by s 43. This difference in the incidence of the persuasive burden is reasonably subtle but is perhaps a reminder of the need for caution in this area.
All in all we are left with the view that there is little or no practical difference between the s 8 and s 43 balancing tests.
Was the evidence about the van incident admissible?
The incident in the context of the case as a whole
As we have noted, Tahani was seen by a Plunket nurse on 15 November 2007. Presumably there was nothing untoward about her condition at that time. The events which are critical to this case thus appear to have occurred over the subsequent six-week period which finished when Tahani was brought into hospital on the morning of 28 December. During those six weeks:
Tahani became severely malnourished;
there were two separate incidents between late November and early to mid-December 2007 in which Tahani suffered serious injuries;
Mr and Mrs Mahomed did not obtain medical assistance for Tahani in relation to those injuries or her malnutrition;
the van incident occurred;
Tahani was badly treated (as acknowledged in the intercepted discussions);
Tahani was fatally assaulted on the night of 27 December 2007; and
in the aftermath of that fatal assault, Mr and Mrs Mahomed did not seek prompt medical treatment.
Events which occurred after Tahani’s admission to hospital included:
collusion between Mr and Mrs Mahomed as to the very similar but obviously false statements which each gave to the police on the night of 28 December 2007; and
subsequent collusive behaviour aimed at coming up with a common narrative as to how Tahani may have suffered her injuries and preventing Tasmia implicating Mr Mahomed.
This pattern of events had probative value in relation to issues in dispute in the proceeding in a number of ways:
The behaviour of Mr Mahomed identified in  pointed to hostility on his part to Tahani and a willingness to take risks with her health and life. This state of mind was material to counts one, two and three. It made the assaults by Mr Mahomed on Tahani more explicable than would otherwise have been the case. As well, the more hostile Mr Mahomed was to Tahani and the less he felt constrained by the likelihood of serious consequences for her, the easier it became to infer that he had the relevant states of mind (intention to cause grievous bodily harm and murderous intent) alleged by the Crown.
There was a very clear interconnection between the grievous bodily harm counts and the murder count. This is because the case against Mr Mahomed was very much based on remarks in the intercepted communications to the effect that he had hit Tahani, remarks which on the Crown case encompassed all assaults.
The behaviours referred to in (a), (b), (c) and (d) were all material to explaining why Mr and Mrs Mahomed did not seek prompt medical treatment on the night of 27 December and the morning of 28 December.
Also material to count four was the subsequent collusive behaviour of Mr and Mrs Mahomed. On the Crown case, this showed that Mrs Mahomed was prepared to cover up for her husband’s offending. Her willingness to do so was very consistent with the Crown case as to the reason why she delayed seeking medical treatment for Tahani.
Finally the van incident was, to the extent to which it was explicable on the basis of guilty knowledge, material to Mr Mahomed’s guilt on the two grievous bodily harm charges.
Although this formulation of the relevance of the evidence is in some respects more detailed than the way the Crown advanced its case, each of these arguments was relied on, to a greater or lesser extent. With the possible exception of the point alluded to in (e), all of these points rested on reasoning along the lines that the relevant behaviour of Mr and Mrs Mahomed showed that each tended to act and think in a particular way towards or about Tahani. But the argument for the appellants in this Court focussed only on evidence about the van incident. There was no challenge to the admissibility of the other evidence we have referred to.
The probative value of the evidence as to the van incident might be thought to be constrained by the following considerations:
The van incident provided the jury with little or no assistance in determining which of Mr and Mrs Mahomed had inflicted either the fatal injuries (or indeed the earlier injuries). On this part of the case, the Crown rested almost exclusively on the intercept evidence.
The Crown case as to the non-accidental nature of the injuries and mens rea was predominantly based on the nature of the injuries and the degree of force which was required to inflict them.
The case against Mr and Mrs Mahomed on the failing to provide the necessaries of life count was strong even without the evidence about the van incident.
That said, we consider that the probative value of the evidence of the van incident was considerable. Properly considered, its relevance was that it was one event in what the jury was well entitled to conclude was a very distinctive pattern of events which, as a whole, was highly relevant to the case. It is inappropriately atomistic to focus on each of these events individually and look for substantial discrete and independent incremental probative value in the associated evidence. If evidence of the van incident was to be excised, why should the malnutrition evidence not be excluded as well? If the jury was not to hear evidence about the van incident as a specific incident involving mistreatment of Tahani, would there be any justification for evidence of specific and general acknowledgements of mistreatment in the intercepted communications? And if all that went, what would be the logic of treating the evidence as to counts one, two and three as all being cross-admissible?
The difficulty for the appellants is that the prejudicial effect of the van incident evidence was conterminous with its relevance. To put this a slightly different way, the evidence revealed nothing discreditable about Mr and Mrs Mahomed which was not directly germane to the case against them. What was discreditable about the van incident from the point of view of Mr and Mrs Mahomed was that it showed that they were prepared to expose Tahani to risk. Their willingness to do so was a central plank of the Crown case and might be thought to have been well established by other evidence. These considerations leave no scope for concern that the van incident evidence had (or risked having) an unfairly prejudicial effect. In terms of s 43(4)(a), the evidence was unlikely to unfairly predispose the jury against the defendants so that they would not apply the requisite standard of proof. Nor, in terms of s 43(4)(b), do we consider that a jury would tend to place disproportionate weight on the van incident in reaching their verdicts..
One of the complaints made by counsel for the appellants was that in the context of the case as a whole, the amount of trial time devoted to the van incident was disproportionate to its value. We accept that if the other alleged misconduct is disputed, proportionality considerations may be engaged given the associated lengthening of the trial and the risk of the jury becoming distracted by what, in the end, may simply be satellite issues. So sometimes the cake may not be worth the candle. Where proportionality is in issue, counsel and judges should consider whether there is a way round what would otherwise be the need for lengthy evidence to be called. One way might be for an agreed statement of facts to be given to the jury. In the present situation however we do not see a proportionality problem; this given (a) the very short period of time (around six weeks) focused on by the prosecution, (b) the pattern of apparently interconnected events (of which the van incident was one) which occurred during this time and (c) references to the van incident in the intercepted discussions.
The section 43 criteria
The s 43(3) and (4) non-exhaustive lists of criteria were plainly prepared with orthodox similar fact evidence in mind and some are of little or no practical application to the evidence in issue before us. But taking as a point of similarity that the van incident involved Tahani as did the various charges Mr and Mrs Mahomed faced, all the s 43(3) considerations other than those provided in ss 43(3)(d) and (e) might be thought to favour admissibility. And for the reasons given in , the s 43(4) criteria fail to assist the appellants.
In cases of child neglect and homicide, evidence broadly similar to that relating to the van incident has been routinely admitted by New Zealand courts both before (R v Meynell  1 NZLR 507 (CA) and R v Witika (1991) 7 CRNZ 621 (CA)) and after (R v Broadhurst  NZCA 454) the coming into effect of the Evidence Act. There are many cases in Canada where the same approach has been taken. The fundamental reason why such evidence has been admitted in the past is that the courts have recognised that its probative value exceeds the risk of an unfairly prejudicial effect on the defendant. The reasons why this is so have not changed with the enactment of the Evidence Act 2006 and are, of course, of direct materiality to the balancing evaluation required by s 43(1).
Conclusion as to admissibility
For the reasons given, the evidence was admissible.
Did the Judge’s treatment of the van incident evidence give rise to a miscarriage of justice?
The New Zealand approach to mandatory jury directions
The appropriate approach to whether particular types of jury direction must be given was explained by this Court in Wi v R  NZSC 121,  2 NZLR 11 in the context of the supposed requirement to give a particular good character direction where the defendant has no convictions. This is what the Court said:
It must be obvious to a jury that evidence that the defendant has no previous convictions is adduced for the purpose of suggesting that he or she is thereby the less likely to have committed the offence charged. Defence counsel will almost certainly have addressed them to that effect. It hardly needs the judge to remind them of the point. To be balanced, a reminder might well have to be accompanied by an invitation to compare the weight (stated implicitly or expressly to be slender) of this evidence with the weight the jury finds in the evidence adduced by the Crown. Such a direction might not be particularly helpful to the defendant and, in any event, the jury might consider that the judge, by directing on such a small and obvious point, was somehow suggesting it had more force than it deserved.
We are of the view that in principle mandatory directions should be reserved for cases in which they are essential to ensure the defendant has a fair trial. It is generally better to leave the extent and content of directions to the trial judge who has the feel of the case. We recognise it was said in Falealili that this approach might tend towards uncertainty and inconsistency. We are not, however, persuaded that this is likely to be a significant problem.
Summings-up should be tailored to the particular case. Specimen directions are helpful when they are required. A very important part of a trial judge’s function is to give the jury as much help as possible by identifying the issues presented by the case and the evidence which is relevant to those issues. The summing up should also, to the extent necessary, explain to the jury in what way evidence should and should not be used. Directions should not be mandatory unless, without them, there is a real risk that the jury will approach the matter in an inappropriate way or in a way which does not do the defendant’s case justice.
New Zealand appellate courts have thus been less prescriptive than their Australian and Canadian counterparts as to how trial judges should sum-up to juries and, in particular, far less enthusiastic about requiring particular forms of direction to be given in respect of commonly recurring issues.
General treatment of propensity evidence
Consistently with the Wi approach New Zealand appellate courts have seen propensity or similar fact evidence as warranting careful treatment by trial judges. The reasons why are obvious. Commonly, propensity evidence relates to misconduct by the defendant with people other than the victim and in respects which are extraneous to the alleged offending. The jury is thus likely to conclude that the defendant is of bad character but the relevance of this to the case at hand will usually be indirect and perhaps of limited weight. The true relevance of the evidence may not be apparent to jurors. As noted, the cogency of such evidence usually turns on ideas about coincidence and probabilities but the associated principles of probability theory are likely to be unfamiliar to most jurors. There may also be a risk that the jury may seek to reason directly from a conclusion that the defendant is of bad character to a finding of guilt. This risk will be enhanced if the true and legitimate relevance of the evidence is either not obvious or not explained to the jury. There is thus usually the double danger that a jury, if not assisted by the Judge, will miss the true relevance of the evidence and instead reason from it in an unsafe way, perhaps in terms of general propensity.
For these reasons, in New Zealand and in like jurisdictions, where propensity (or similar fact) evidence is adduced, there is usually a requirement for trial Judges to identify the relevance of the evidence, and to warn the jury of its limitations, in particular against engaging in illegitimate reasoning processes. It is important to realise, however, that the requirement for such directions is not based on anything which appears in the Evidence Act 2006. Rather it is one which appellate courts, operating under statutory provisions equivalent to s 385(1)(c) of the Crimes Act 1961, have imposed on trial judges.
The Stewart direction
In Stewart (Peter) v R  NZCA 429,  1 NZLR 197 (at ) the Court of Appeal noted that a propensity direction should explain to the jury what relevant evidence has been called, to what issue it is directed and how the jury may use it. It also, in adopting the argument of counsel for the appellant, indicated that a judge summing up on propensity should adopt a seven-stage process:
State the purpose of the witness’ evidence.
Explain what the propensity evidence is. While it is defined in s 40(1)(a) (set out above at para ), “propensity” is a term with which jurors may be unfamiliar. It would be helpful to give the dictionary definition of propensity as from the Shorter Oxford English Dictionary cited in R v Tainui  NZCA 119 at para .
Identify the factors relied upon by the Crown establishing the relevant propensity in relation to the issue identified at step (1) and referring to any defence contentions. It was agreed by the parties that a judge should refer the jury only to such specific considerations in s 43(3) as are relevant. We agree. While the presence of certain of the factors may be relevant both to the judge’s decision as to admissibility and to the jury’s as to their verdict, reference to what is not directly relevant would act as a distraction from their task.
Direct the jury that whether propensity existed is entirely a matter for them.
Explain that, if propensity is found to exist, it is to be used as circumstantial evidence to be considered with all other evidence when assessing the issues, including the reliability and credibility of the complainant.
Explain that, if the jury does not accept that propensity is properly established it should put the evidence outside and leave it entirely out of consideration.
Warn the jury not to jump to a conclusion that because the accused has offended on a previous occasion he must have done so in a manner alleged in the charge: R v Taea  NZCA 472 at .
As will become apparent we see some difficulties with this seven-step process. It is likely to prove unnecessarily complex and at least in some contexts may confuse rather than enlighten the jury. In this respect it is important to recognise that given our broad and literal approach to the definition of propensity evidence, much more evidence is encompassed by it than was perhaps thought to be the case when Stewart was decided. In particular, evidence which is within the s 40 definition of propensity evidence may be led for reasons which primarily do not invoke orthodox similar fact reasoning.
Some reasons why propensity evidence might be led and the associated risks of unfair prejudice
Professor JR Spencer has identified three different types of propensity evidence:
“Scenario one”, where the Crown can establish by practically irrefutable evidence that the defendant has a particular propensity. An example might be where a male defendant is charged with sexual offending against a female child and the Crown adduces either evidence of convictions for paedophile offences or photographs of the defendant committing such offences. In this situation, the Crown will rely on the defendant’s established propensity to support its case. Sometimes the logic invoked is that the propensity is akin to an identifying characteristic. Thus in the example just postulated, the fact that the defendant is a paedophile provides direct support for the complainant’s evidence, in the sense that he happens to have the particular characteristic which she attributes to him, and thus increases the probability of her evidence being true. But underpinning, or at least closely associated with, this approach, will usually be coincidence reasoning (whether articulated or not). So, staying with the postulated example, the Crown might stress the implausibility of a complainant who wishes to make a false complaint happening to choose someone who just happens to be a paedophile. The well known case of R v Straffen  2 QB 911 (CA) illustrates the point we have just made. This case concerned the murder of a young girl in very particular circumstances and in close proximity to the Broadmoor Institution. The murder occurred in a four-hour period during which the defendant, a Broadmoor inmate, was at liberty following an escape. The Crown could establish that he had previously killed two other young girls in very similar circumstances. The reality was that either the defendant was the murderer or there were, at the time of the murder and in close proximity to Broadmoor, two people with a very particular and distinct character trait involving the willingness to kill young girls.
“Scenario two”, where the evidence is simply circumstantial. Professor Spencer’s example of this is R v Wallace  EWCA Crim 1760,  1 WLR 572, where the defendant was linked by some circumstantial evidence to no less than four robberies, all of which had similar features. Other examples which come to mind are the well-known baby farming cases (Makin v Attorney-General for New South Wales  AC 57 (PC) and R v Dean (1895) 14 NZLR 272 (CA)) and the “brides in the bath” case (R v Smith (1915) 11 Cr App R 229 (CA)). In such a case, the relevant propensity attributed to the defendant is not really a stepping stone on the way to a conclusion that the defendant is guilty. Instead the conclusion that the defendant has such a propensity is essentially a corollary – and thus down-stream – of a finding that the charges have been made out. In saying this we recognise that where the evidence of guilt in relation to one or more of the incidents is much stronger than in relation to the others, scenario three is similar to scenario two.
“Scenario three”, where a number of witnesses give disputed evidence of broadly similar offending (usually of a sexual nature) by the defendant. The evidence of each witness supports that of the others because of the unlikelihood that independent witnesses would make up similar stories. If the evidence of one of the witnesses is much stronger than that of the others (perhaps because of independent corroboration), scenario three is similar to scenario one. But more commonly, as with scenario two, a conclusion that the defendant has the relevant propensity will simply be a corollary of, and not a stepping stone to, a conclusion that the charges have been proved.
As is apparent from what we have just said, in the very usual situation where the evidence may be more compelling in relation to one incident than the others, scenario two and three cases may be susceptible (depending on the view the jury takes of the evidence in relation to each incident) to a scenario one classification. And likewise, if in what the Crown treats as a scenario one case, the evidence of the alleged propensity turns out to be not overwhelming, the case may well have to be looked at as a scenario two or three case.
Although scenario one cases can, at least usually, be analysed in terms of coincidence, it will sometimes be better to explain the relevance of the propensity evidence as being material, in a specified way to the probability of the defendant being guilty.
Where the relevance of the evidence turns primarily on coincidence reasoning, there is a need for careful consideration of the plausibility or otherwise of the postulated coincidence. Where the victim of sexual assault has initially identified the assailant from a set of photographs of local sex offenders, the fact that the person identified has convictions for sexual offences does not add much weight to the identification (because anyone identified from those photographs would have a similar background). If instead the victim, by chance and thus without pre-arrangement, identified the assailant in the street, that person’s prior convictions for similar offending might be thought to provide considerably greater support to the identification.
Professor Spencer’s scenario one involves the greatest and most obvious risk of unfair prejudice because the jury will inevitably conclude that the defendant is a person of bad character whereas in simple scenario two and three cases a conclusion that the defendant is a person of bad character will usually be a consequence of a conclusion that the defendant is guilty rather than a stepping stone to that conclusion. Thus in R v Wallace (the case already mentioned where the appellant was linked circumstantially to four similar robberies), the Court of Appeal took the view that the case did not turn on whether the defendant had a propensity to commit offences but rather the jury’s overall assessment of all the circumstances. For this reason the trial Judge had not been obliged to give a bad character direction. There is other authority to the same general effect. It is, however, right to recognise that there may be a particular risk of unfair prejudice where the evidence of guilt in relation to one of the events is stronger than in respect of the others. Because it will often be unclear how the jury is going to assess the strength of the evidence directly referable to each incident, it will generally be prudent to assume that there is a risk of unfair prejudice in all three scenarios.
The three scenarios just discussed do not address propensity evidence as it relates to interactions between the defendant and victim. In respect of this:
The propensity evidence may be relevant for reasons associated with coincidence, such as the implausibility of a young child receiving a number of injuries by accident. In a broad sense, this situation is similar to Professor Spencer’s scenario two.
The propensity evidence may have important explanatory value, as bearing on the background or relationships between those involved in or affected by the alleged offending.
As a subset of (b), the propensity evidence may be relevant to establishing hostility on the part of the defendant to the victim or a motive for the defendant to harm the victim.
As a further subset of (b), events may be so interconnected with the offending that the jury will not be able to understand properly what happened without hearing evidence about those events.
Leaving aside the first of the situations just postulated, in these circumstances the Crown will not usually be much reliant on ideas about coincidence and probability and the wrongfulness of the defendant’s conduct will usually be so closely connected to the core elements of the case against the defendant as to leave little scope for unfair prejudicial effect.
When is a propensity evidence direction required?
A propensity evidence direction is required where the Crown is:
relying on propensity reasoning and in doing so is invoking ideas about coincidence or probability; and/or
the evidence involves aspersions on the character of the appellant in respects not directly associated with the alleged offending.
As well, a propensity evidence direction should be given where, without it, there is a danger that the jury will not realise the relevance of the evidence in question or there is some particular risk of unfair prejudice associated with the evidence.
On the other hand, and as the corollary of what we have just said, where the evidence in question, although still falling within the Act’s “propensity evidence” definition, is not led primarily in reliance on coincidence or probability reasoning, a specific direction may well not be required.
As to this last point, our impression of the New Zealand jurisprudence referred to in  above is that in the past particular directions were not usually given, nor seen as a requirement, in relation to evidence of misconduct by the defendant towards (R v Rongonui  2 NZLR 385 (CA) is an exception) victim. Canadian cases which have explicitly addressed this issue support the view that particular directions are not required in relation to evidence as to hostility between the defendant and victim. Prior to the enactment of the Criminal Justice Act 2003 (UK) – which introduced a new regime in relation to bad character evidence – the English courts usually saw evidence of misconduct by the defendant to the victim as not engaging the similar  rules, with the result that particular directions were not required.
How should a propensity direction be constructed?
Our far from complete analysis of the dispiritingly various ways in which propensity evidence might be relied on shows that there is no scope for a one-size-fits-all standard propensity direction. And although the Court of Appeal in Stewart rightly recognised the need for flexibility and the need to tailor directions to the facts of the particular case, its attempt to provide something of a universal template was overly ambitious. As well, there are some components in the seven-step process which we think are likely to be unhelpful from the point of view of a jury.
When giving a propensity evidence direction a judge should:
Identify the evidence in question and explain why it has been led and the legitimate respects in which it might be taken into account by the jury. We see no need for the judge to define “propensity” (compare step (2) of Stewart). In cases in which a demonstrated propensity could legitimately be a stepping stone in the reasoning process of the jury, that should be identified using concrete language addressed not to “propensity” as an abstract concept, but rather specifically to the particular pattern of behaviour or thinking which is in issue. In most cases, the legitimate reasoning available to the jury will be based around coincidence or probability. That should be explained to the jury in simple and direct language addressed to the particular facts and what is said to be the implausible coincidence or how the evidence otherwise bears on the probability of the defendant being guilty. This is likely to require a discussion of the similarities involved in the conduct alleged. Where there are factors which may explain the postulated coincidence (for example, suggested collusion between the witnesses) that too should be addressed. We see no need for the judge to otherwise go through the s 43(3) criteria (compare step (3) of Stewart). These criteria are addressed to the admissibility decision the judge must make and not the factual assessment which is for the jury.
Put the competing contentions of the parties.
Caution the jury against reasoning processes which carry the risk of unfair prejudice associated with the propensity evidence. This should usually be along the lines that the fact that the defendant has or may have offended on other occasions does not establish guilt and that the only legitimate reasoning process available to the jury is the one which has been outlined.
The case at hand
In this case, the Judge did not specifically direct the jury as to how they might use the van incident evidence and he did not give any particular direction of the kind usually given in relation to propensity evidence. Apart from warning the jury against being influenced by prejudice or emotion, all he did was paraphrase the principal contentions of counsel. In issue before us is whether this led to a miscarriage of justice.
When counsel asked the Judge after his summing-up to give a propensity direction, it was by specific reference to the style of direction approved in Stewart. The Judge’s refusal to direct in that way is unsurprising as such a direction would have been over-complex and unhelpful.
A particular direction confined to the van incident evidence and its relevance and warning against unfair reasoning would have been artificial, given the extent of the evidence before the jury which was within the propensity rule as we interpret it. The relevance of the van incident is that it was one event in what the jury could fairly consider to be a related series of events. A propensity direction addressed to the relevance of the van incident would thus have involved a discussion of the relevant pattern of events along the lines of what appears above in .
It is difficult to see how Mr and Mrs Mahomed would have been advantaged by such a direction. Indeed had it been given it would have made rather more of the propensity arguments than the prosecutor had endeavoured to do in closing and, in the context of the case as a whole, would therefore have been likely to have resulted in the van incident evidence assuming much greater prominence in the thinking of the jury than was the case with the way the Judge actually summed up.
Was there any danger that the jury, unassisted by the Judge, might not have understood the relevance of the van incident?
Because the van incident evidence was directly associated with the attitudes of Mr and Mrs Mahomed to Tahani and to each other, attitudes which on the Crown case provided the context – and indeed an explanation – for the offending, the evidence could fairly be regarded as directly relevant to the offending, in effect as just one component of a circumstantial case. That this was so must have been apparent to the jury. It was just a matter of common sense and there was thus no difficulty associated with relevance which required explanation.
Were Mr and Mrs Mahomed prejudiced by the absence of a caution along the lines mentioned in (c)?
That they had not been good parents was established by a good deal of evidence, including what was said in the intercepted communications. That this was so was perfectly apparent and acknowledged by their counsel. The general way in which they had acted as parents and their associated attitudes to Tahani were a central part of the Crown case and of direct relevance to the counts they faced. It was thus in no way extraneous to any of the alleged offending. As noted, the relevance of the evidence was obvious. So there is no reason to think that the jury may have engaged in illegitimate reasoning as a result of an inability to understand why the evidence was legitimately material.
A warning not to jump to the conclusion that Mr and Mrs Mahomed were guilty just because they were bad parents would have been a statement of the obvious with the result that the Judge’s failure to give such a warning is of no moment. This is particularly so given his paraphrase of what was said by defence counsel on this point; see above at . The Judge’s warning not to be affected by prejudice or emotion must have had substantial resonance with the jury given the facts of the case and there is no reason to think that they did not act on it.
In this regard, remarks made by Lord Phillips of Worth Matravers CJ in R v Campbell  EWCA Crim 1472,  1 WLR 2798 warrant attention. He referred to the difficulties which had arisen in the past when common sense directions as to character evidence had not been given resulting in the quashing of convictions. He then went on:
This lamentable state of affairs kept the Court of Appeal and, not infrequently, the House of Lords, busy with appeals against jury directions in respect of character. It is an extreme example of the way that directions that are desirable by way of assisting the jury to draw logical conclusions from the evidence have become treated as if they are mandatory requirements of law.
This often occurs in the following way. The Court of Appeal criticises an aspect of a judge's summing up and suggests an alternative direction that would have been appropriate. The Judicial Studies Board then incorporates this suggestion in a specimen direction. Thereafter, if the specimen direction is not given, this is treated as a defect in the summing up that warrants permission to appeal and has, on occasion, been treated in this court as rendering the conviction unsafe without considering whether the jury would have reached the same conclusion by the application of common sense to the evidence, whether or not the specimen direction was given. Failure to give a direction that is no more than assistance in applying common sense to the evidence should not automatically be treated as a ground of appeal, let alone as a reason to allow an appeal.
We agree and consider that the emphasised passage is directly relevant to the present appeals. While the summing-up was short on detail, the relevance of the evidence was obvious and the assistance which the Judge gave the jury was adequate.
Accordingly we are of the view that the way in which the Judge summed up to the jury did not result in the trial miscarrying.
For those reasons the appeals should be dismissed.
 We agree with William Young J’s reservations about the utility and content of the directions suggested in Stewart (Peter) v R  NZCA 429,  1 NZLR 197, albeit they were not proffered as mandatory. We prefer to leave further consideration of those directions to a future case in which it is required.
 And which refers to the prejudicial effect on “the defendant”, rather than “the proceeding”, the significance of which is discussed below from .
 See below at –
 See for instance R v Howse  UKPC 30,  1 NZLR 433 at –, ; R v Howse  3 NZLR 767 (CA) at –.
 Because the conduct in question was between defendants rather than between a defendant and a victim.
 It primarily related to them sharing a bedroom in which there was only one bed.
 This is the illustration given by Binnie J in R v Handy 2002 SCC 56,  2 SCR 908 at 939.
 That is, in respects quite independent of the alleged offending against the victim. This is a point made by the Ontario Court of Appeal in R v Batte (2000) 49 OR (3d) 321 at – and discussed by David M Paciocco and Lee Stuesser The Law of Evidence (5th ed, Irwin Law, Toronto, 2008) at 69–70.
 R v Accused (CA 247/91)  2 NZLR 187; R v Witika (1991) 7 CRNZ 621 (CA) and R v Macdonald CA166/04, 8 April 2005, where the issue is reasonably fully discussed.
 In the Australian High Court decisions (see Gipp v R (1998) 194 CLR 106 and HML v R  HCA 16, (2008) 235 CLR 334), there is a particular focus and concern about uncharged acts. But logically, the admissibility of evidence about other offending by the defendant towards the victim cannot depend on whether the defendant also faced charges (and is being simultaneously tried) in relation to the other alleged offending.
 See the discussion below at –.
 See for instance R v Roud (1981) 58 CCC (2d) 226 (Ont CA); R v Schell (1977) 33 CCC (2d) 422 (Ont CA) and R v Kematch 2008 MBQB 277.
 As explained by JR Spencer Evidence of Bad Character (2nd ed, Hart Publishing, Oxford, 2009) at [1.27]–[1.30].
 In the UK see R v Hanson  EWCA Crim 824,  1 WLR 3169; R v Campbell  EWCA Crim 1472,  1 WLR 2798. In Australia see R v Beserick (1993) 30 NSWLR 510 (CCA); Gipp v R (1998) 194 CLR 106 and the discussion in JD Heydon Cross on Evidence (looseleaf ed, LexisNexis Butterworths) at . In Canada see R v B (FF)  1 SCR 697, the discussion in Alan W Bryant, Sidney N Lederman and Michelle K Fuerst Sopinka Lederman and Bryant: The Law of Evidence in Canada (3rd ed, LexisNexis, Canada, 2009) at [11.181]–[11.186] and Paciocco and Stuesser The Law of Evidence at 76–77.
 Set out at .
 JR Spencer Evidence of Bad Character (2nd ed, Hart Publishing, Oxford, 2009) at [4.67]–[4.70].
 This is helpfully discussed by Mike Redmayne “Recognising Propensity”  Crim LR 177 at 182–193; David Hamer “The Structure and Strength of the Propensity Inference: Singularity, Linkage and the Other Evidence”  Monash U Law Rw 7; (2003) 29 Monash University Law Review 137 at 157–163, and Rosemary Pattenden “Similar Fact Evidence and Proof of Identity” (1996) 112 LQR 446 at 454–463.
 This and related difficulties (for instance as to jury directions) are usefully discussed by Mike Redmayne “Recognising Propensity”  Crim LR 177 at 188–192.
 For instance by showing that the defendant on other occasions has behaved in broadly the same way as the complainant alleges, propensity evidence might be thought to add to the plausibility of that evidence. Hudson v R  NZSC 51 provides an example of a situation where directing the jury on particular propensity evidence in terms of coincidence would have been unhelpful.
 A well-known example where there was a failure to interrogate rigorously the alleged coincidence is the case of Sally Clark, see R v Clark  EWCA Crim 1020,  2 FCR 447. Mrs Clark was convicted of the murder of her two very young children after a trial in which the Crown led significantly misleading statistical evidence suggesting that the chances of two children in the same family dying of sudden infant death syndrome were one in 73 million. For a sequel to this case, where the flaws in the statistical evidence are closely dissected, see General Medical Council v Meadow  EWCA Civ 1390,  QB 462.
 A consideration which was discussed in R v Broadhurst  NZCA 454. Coincidence reasoning could perhaps have been, but was not, invoked in the present case.
 Although coincidence reasoning may be relied on to some extent. Thus where the defendant is alleged to have murdered the victim, prior actions by the defendant to the victim may be primarily relied on to establish motive as a direct component of the case against the defendant. The Crown, however, might also say that the defence hypothesis that the defendant is innocent involves the unlikely coincidence of two people having had a motive to kill the victim.
 For the reason discussed above in .
 See R v Krugel (2000) 143 CCC (3d) 367 (Ont CA), R v Holtam 2002 BCCA 339, (2002) 165 CCC (3d) 502 and R v Merz (1999) 46 OR (3d) 161 (Ont CA).
 See the note by Professor Birch “Evidence: murder - admissibility of previous violence by defendant to deceased”  Crim LR 651; R v M  1 WLR 421 (CA) at 426–427 and PJ Richardson (ed) Archbold: Criminal Pleading Evidence and Practice (2002 ed, Sweet and Maxwell, London) at [13-34].
 The evidence we have in mind is set out above at –.
P L Borich(Rice Craig, Papakura) for Appellant Tabbasum Mahomed.
G J King and C B Wilkinson-Smith for Appellant Azees Mahomed.
M D Downs and K A L Bicknell (Crown Law Office, Wellington) for Crown.
all rights reserved