New complications for old

Mahomed v R [2011] NZSC 52 (19 May 2011) shows that the law of similar fact evidence is still contentious. It has recently been reformed, by calling it propensity evidence and by enacting it. Sections 40-43 of the Evidence Act 2006 [NZ] replace the common law on this topic.

In Mahomed the Supreme Court split 3-2 over what items of evidence were included as propensity evidence. The majority focused narrowly on one incident, and the minority said it was artificial to take an “atomistic” approach [73] and instead the overall pattern of events had to be looked at. The same split occurred on whether the propensity evidence was admissible: the majority said the one incident was only admissible on one of the four counts, whereas the minority said the overall pattern of events was admissible on all counts.

Further, the judges split into the same groups on whether it was appropriate to consider in detail the kinds of directions that juries should be given. The majority chose not to go into this, whereas the minority analysed the use and risks of propensity evidence, when a direction should be given and what it should say.

Conveniently, the consistency of the split between the members of the Court made it only necessary to deliver two judgments. The majority judgment of Elias CJ, Blanchard and Tipping JJ was delivered by Tipping J, and for the minority William Young J delivered the judgment of himself and McGrath J.

As the majority found there had been a miscarriage of justice arising from the way the judge had instructed the jury (he had followed a Court of Appeal ruling in this case – R v Mahomed [2009] NZCA 477 – which had held that the evidence was admissible on all counts), it had to consider whether to apply the proviso. As it happened, doing this by applying R v Matenga [2009] NZSC 18 (an approach I have criticised here, for example on 20 July 2009 and 1 January 2010) involved no greater discussion of the prosecution case that would the more conventional method for deciding conviction appeals which asks whether there was a real possibility that the error at trial affected the verdict (see Fraser v HM Advocate [2011] UKSC 24, discussed here on 26 May 2011).

The Mahomed minority’s analysis of types of propensity evidence, identifying categories and sub-categories, and allowing that some instances may require swapping between categories depending on the strength of the evidence, leaves one wondering if the subject is set to become, once more, over-academicised. Complications arising from “coincidence” reasoning (eg at [51]) tend to confuse rhetoric with substance. When a prosecutor argues that either the defendant is guilty or that he is the victim of an amazing coincidence through being falsely accused by independent complainants, that rhetoric is really a reflection of the denominator of the Bayesian likelihood ratio: the low probability of getting the accusations on the assumption that the defendant is innocent.

The prosecutor’s argument is sound, but it does not generate a new category of propensity evidence (see [85(a)], and [87] where Hudson v R [2011] NZSC 51 is said in footnote 50 not to be an appropriate case for coincidence reasoning). But coincidence is just another word for “low probability” and as para [59] of Hudson points out coincidence reasoning there was unnecessary as it would have duplicated (“elaborated”) what the judge had told the jury. The prosecution would have said there that as the defendant had a motive, commission of the offence by someone else would by an unlikely coincidence. It comes down to the low probability of getting the evidence of motive on the assumption of innocence, and the high probability of getting it on the assumption of guilt.

There was unanimity on the relationship between the balancing exercises in s 8 and s 43 which both address probative value and prejudicial effect. In the context of propensity evidence, s 43 is the sole balancing exercise: majority at [5], minority at [66-67]. The minority do, however, accept a theoretical case of equal balance, but I think that is a misconception. The balancing metaphor in both sections does not work like a weighing of commodities. What needs to be avoided is a real risk of unfairness to the defendant, and probative evidence can “outweigh” that risk if admission of the evidence does not give rise to that risk. The Court of Appeal recognised this in Vuletich v R [2010] NZCA 102 per Glazebrook J at [27] and Randerson J at [96], rejecting a suggestion that had been made in Stewart (Peter) v R [2008] NZCA 429 by Baragwanath J to the effect that highly probative evidence could only be inadmissible if it gave rise to a level of illegitimate prejudice that was even greater. It would have been useful for the Supreme Court to have addressed this fundamental issue in Mohamed.


Non-disclosure and trial fairness

An appellate court often has to ask whether a trial was fair. Statutory requirements for allowing conviction appeals usually include “a miscarriage of justice”. What does that phrase mean for the purposes of assessing whether the trial was fair? The United Kingdom Supreme Court addressed this in Fraser v HM Advocate [2011] UKSC 24 (25 May 2011).

In this case information had not been disclosed to the defence, and had it been disclosed the conduct of the case for each side would have been different. Therefore, non-disclosure here went to the question of trial fairness [32].

The question of trial fairness is not answered by considering what verdict the appellate judges would have reached at a hypothetical trial [38]:

“One cannot, of course, avoid making some assumptions as how the trial might have been conducted if the material had been disclosed to the defence. It will always be a question of degree as to how far it is proper to go in carrying out that exercise. But the purpose of doing this is to assess the extent to which, having regard to the way the case was conducted by the Crown, the material would have weakened the Crown case or strengthened the case for the defence. It is on the case as presented at the trial that the court must concentrate, rather than the case as it might have been presented. It is not for us to speculate as to what the case might have been, much less how the jury would have reacted to it. What the Crown asks us to do, and what it persuaded the Appeal Court to do, was to consider the case on the basis that the discovery of the rings on 7 May was indicative of the appellant’s guilt for completely different reasons from those advanced at the trial. In effect we were being asked to deal with the case as if we were a new jury trying the case for the first time. This is not permissible. Our task is quite different but entirely clear. As the Appeal Court said in McCreight v H M Advocate [2009] HCJAC 69, 2009 SCCR 743, para 95, it is not the court’s task to decide what the outcome of the trial would have been if the trial had been conducted on an entirely different basis. We must ask ourselves whether, in the light of the undisclosed evidence, there is a real possibility that the jury at this trial would have arrived at a different verdict.”

The trial would have been unfair if there was a real possibility that had the error not occurred the jury would have reached a different verdict.

The case illustrates how an appellate court should apply this test. At trial the prosecution had alleged that the defendant’s guilt of his wife’s murder was demonstrated by her rings that could only have been found where they were if he had removed them from her corpse. The non-disclosed evidence was that the rings had been at that location before she died.

“[39] The proposition that the appellant had returned the rings to the bathroom on 7 May was, as the Advocate Depute said in his address to the jury, the cornerstone of the Crown’s case. It is clear, in view of the direction that was then given to them by the trial judge, that the jury must have concluded that the appellant put the rings in the bathroom on 7 May. This was the basis for the Crown’s theory that he had obtained the rings from the deceased’s dead body and had placed them in the bathroom to create the impression that she had left the matrimonial home with the intention of turning her back on the life that she had had there. This theory would have been undermined by the evidence of PC Lynch and WPC Clark. It would have been challenged by lines of cross-examination of the Crown witnesses that were never developed at the trial, and by questions that were never put to the appellant in chief or in re-examination. The point could have been made that it was improbable that, if the rings were in the bathroom on 28 and 29 April when the police visited the house, the appellant would have removed them and then chosen to return them on 7 May. The theory that he removed them from the dead body would, if the evidence of PC Lynch and WPC Clark were to be accepted, have been untenable. These and other arguments that the defence would have been able to develop would have struck at the heart of the case that the Crown presented. The trial would have been significantly different had the material that was not disclosed been available. There is a real possibility that this would have been sufficient to raise a reasonable doubt about the Crown’s case that the appellant returned the rings to the bathroom on 7 May. If that were so, the jury’s verdict would be bound in view of the trial judge’s direction to have been different.”

It was the real possibility of a different verdict that made the trial unfair.

Non-disclosure significantly affected the course of the trial. This was not a case of freshly discovered evidence, although I think the difference should be immaterial because the ultimate question is whether there is a real possibility that the omission affected the verdict.

I have suggested that all conviction appeals raise issues of trial fairness. This is because a fair trial is one where the law is properly applied to facts that are determined impartially. “Impartially” is used in a wide sense to include actual and apparent bias as well as inaccuracy. Even appeals challenging jurisdiction raise fairness in this sense, as without jurisdiction the law is not properly applied. Fresh evidence appeals also concern fairness, as the newly discovered evidence may demonstrate a real possibility that the verdict was wrong. There can be unfairness even if the verdict was not affected, where the law was not properly applied, but this was not that sort of appeal.

I have recently (7 April 2011) mentioned misuse of hypotheticals by appellate courts, and have also criticised the usurpation of the jury role by appeal judges, for example here (9 July 2009) and here (25 June 2007), and here (16 January 2006).

Writing for judges

Do not even consider living a day longer without reading these interviews with the Justices of the United States Supreme Court: The Scribes Journal of Legal Writing (2010) vol 13, available here.

Because brevity wins, I leave it at that, except to mention this teaser by Roberts CJ:

“What the academy is doing, as far as I can tell, is largely of no use or interest to people who actually practice law.”

Avoiding the paperwork

The United States Supreme Court has handed police what may seem to be a legal way of searching homes without the need to obtain warrants: Kentucky v King USSC No 09-1272 (16 May 2011). When should a defendant’s suspicious behaviour be ignored when the adequacy of grounds for a warrantless search is being assessed by a court?

The fundamental requirement is that there should be reasonable grounds for a search. The police need not apply for a search warrant, even if they have grounds to obtain one. The opinion of the Court, delivered by Alito J and joined by Roberts CJ, Scalia, Thomas, Kennedy, Breyer, Sotomayor and Kagan JJ (Ginsburg J delivered a dissenting opinion), mentioned five reasons for not requiring a warrant. None of these are convincing. They are set out at p 12 of the slip opinion. I summarise:

    1. The police may wish to speak to the occupier to see if it is worth getting a warrant.
    2. The police may want to ask the occupier for consent because that would be easier.
    3. They may want more evidence before submitting a marginal application for a warrant.
    4. They may want evidence to justify a broader warrant than they currently have grounds for.
    5. They may not want to disclose that they have grounds for a warrant because that might alert other suspects.

These were held to be legitimate law enforcement strategies. There will, I acknowledge, be occasions when resort to them is reasonable. It may be absurd to require the police to hold off conducting a search while they apply for a warrant, as indeed it would have been on the facts of the present case. However, the danger is that the safeguards involved in the warrant procedure will be cast aside in favour of expedience on the basis of an imagined urgency. It would be easy for the police to claim that they heard what seemed to be an attempt to destroy evidence, and a citizen subject to warrantless search of his home would be unlikely to be able credibly to challenge its legality. It would be inappropriate for a judge to take any but a very cautious approach to evaluating the credibility of police claims of such urgency. 

Nor is it appropriate to treat the police as if they were private citizens, as the majority do at p 16. Private citizens are not state agents collecting evidence to prosecute suspected offenders. It is disingenuous for the Court to hold that occupiers may “stand on their constitutional rights” and not answer the door or allow police entry, but if they “elect to attempt to destroy evidence [they] have only themselves to blame for the warrantless exigent-circumstances search that may ensue.”

One would think that, on this approach, a refusal by an occupier to allow police entry would be understood by the police to be an attempt to preserve an opportunity to destroy evidence, and would therefore be justification for a warrantless entry and search.

The opportunities for abuse of powers in this context are such that, for people who regard the Fourth Amendment as a proper restraint on executive power, Ginsburg J’s dissent may be considered the sole voice of reason in this case.

But the case is not about random door-knocking by the police. The police believed that the smell of cannabis was coming from a specific apartment.

The ratio of this case is that in deciding whether a warrantless search was reasonable the court will need to consider all the circumstances, and unreasonable police conduct does not give rise to reasonable grounds. This is unobjectionable as a legal rule. If it had been unreasonable for the police to knock on the door in this case, the police would not have been able to rely on subsequent inferences they drew as to risk of destruction of evidence (the exigency). There was no evidence here (pp 17-18) that the police had acted unreasonably prior to entry, so (p 19) the exigency justified the warrantless entry. This was on the Court’s assumption, for the purposes of argument (p 17), that there was in fact an exigency here. This assumption was only necessary if, without the exigency, there would have been inadequate grounds for entry. So the police here had reasonable grounds for knocking on the door but, at that point, not for entry, but when (if) the exigency occurred they had (would have had) reasonable grounds for entry. The case was remanded to the Kentucky Supreme Court for further proceedings not inconsistent with the Court’s opinion.

A strange thing about this case is why exigent circumstances were relevant. If the police smell cannabis (“marijuana”) smoke and establish that it is coming from private property, they would have reasonable grounds to believe that an offence was being committed and so could search without warrant: for example, R v Gurnick [1999] NZCA 19, (1999) 16 CRNZ 513 (CA). But the position is different in the US, where there is a stronger preference for grounds to be assessed by a neutral official: Johnson v United States, 333 U.S. 10 (1948). In that case the smell of opium from a hotel room, without the risk of a suspect fleeing, or of destruction of evidence, was not grounds for a warrantless search.

Counter-intuitive evidence: should neutrality be sought?

Correction of wrong intuitions is important to prevent jury bias. Commonly held assumptions about the way a complainant should behave may unfairly affect the jury’s assessment of that person’s credibility.

Delay in complaining, or the continuation of a relationship between the complainant and the defendant, might be thought to undermine the credibility of the complainant. They may indeed undermine credibility on the facts of a particular case, but it would be wrong for the jury to apply a rule of thumb when assessing the evidence.

In M (CA23/2009) v R [2011] NZCA 191 (18 May 2011) the New Zealand Court of Appeal considered ways of countering the wrong intuitions that jurors may have about the significance of such circumstances to credibility. It referred in particular to decisions of the Supreme Court of Canada, R v DD 2000 SCC 43, [2000] 2 SCR 275, and the Court of Appeal of England and Wales, R v Miller [2010] EWCA Crim 1578. Expert witnesses may give evidence on the counter-intuitive significance of such circumstances, the judge may direct the jury on the dangers of making assumptions, or counsel may agree on a statement to the jury on the point.

Those techniques of countering wrong intuitions are aimed, said the Court in M(CA23/2009) at [25], adopting a comment by the New Zealand Law Commission, at restoring a complainant’s credibility from a debit balance because of jury misapprehension, back to a zero or neutral balance. It needs to be added that the defendant should also be allowed to use these techniques to prevent wrong intuitions being used against him.

We might wonder whether the aim of setting the credibility balance back to neutral is necessarily correct. Some kinds of behaviour may well indicate reliability in general. Not everything is neutral. Delay may be neutral, but a quick complaint may in the generality of cases indicate reliability. The continuation of an apparently good relationship between the complainant and the defendant may be neutral, but a sudden breakdown of a relationship may in general indicate that the complaint is true. Or in turn those intuitions may prove to be false. What do the scientific studies show?

There is no reference in M(CA23/2009) to statistical data on the significance of types of behaviour to the truth or falsity of complaints. But at least we can say that logically it is necessary to compare the occurrence of the behaviour in the cases of true complaints, to its occurrence in the cases of false complaints.

The jury must not be left in the impossible position of having to use their collective common sense while at the same time treating as neutral behaviour that may well be useful in assessing credibility. How can a jury decide the significance of such behaviour in the particular case without putting it in a more general context?

Compensation for wrongful convictions

When should a person whose conviction has been quashed receive compensation? This question of entitlement is different from the question of quantum of compensation. Although it is arguable that the two questions should be considered together, they were not in Adams, R (on the application of) v Secretary of State for Justice [2011] UKSC 18 (11 May 2011).

Two poles are identifiable: one is that people who are actually innocent should receive compensation after erroneous conviction, and the other is that no one who is actually guilty should receive compensation after erroneous conviction. By “erroneous” conviction I mean not in accordance with the law.

The problem is to navigate between these poles when interpreting the criteria for compensation. In this case it was the interpretation of s 133 of the Criminal Justice Act 1988[UK] that governed the issue, in particular the phrase “a new or newly discovered fact shows beyond reasonable doubt that there has been a miscarriage of justice“. The Court split 5-4 on what this should mean.

The dissenters were particularly revolted by the thought that a person who was in fact guilty might receive compensation after erroneous conviction. Lord Brown (with whom Lord Rodger agreed without delivering a separate judgment) said at [281]:

“Why should the state not have a scheme which compensates only the comparatively few who plainly can demonstrate their innocence – and, as I have shown, compensate them generously – rather than a larger number who may or may not be innocent?”

The minority would accept that some people who really were innocent should be denied compensation in the interests of avoiding giving compensation to the actually guilty. Other minority judges were Lord Judge and Lord Walker.

Lady Hale succinctly addressed Lord Brown’s “palpable sense of outrage” [116]:

“Innocence as such is not a concept known to our criminal justice system. We distinguish between the guilty and the not guilty. A person is only guilty if the state can prove his guilt beyond reasonable doubt. This is, as Viscount Sankey LC so famously put it in Woolmington v Director of Public Prosecutions [1935] AC 462, at p 481, the “golden thread” which is always to be seen “throughout the web of the English criminal law”. Only then is the state entitled to punish him. Otherwise he is not guilty, irrespective of whether he is in fact innocent. If it can be conclusively shown that the state was not entitled to punish a person, it seems to me that he should be entitled to compensation for having been punished. He does not have to prove his innocence at his trial and it seems wrong in principle that he should be required to prove his innocence now.”

She agreed (as did Lords Hope, Kerr, and Clarke) with the test formulated by Lord Phillips at [55]:

“A new fact will show that a miscarriage of justice has occurred when it so undermines the evidence against the defendant that no conviction could possibly be based upon it.”

The person does not have to show his innocence beyond reasonable doubt.

Under this test there will be occasions where people who are in fact guilty are entitled to compensation. Such cases may occur where, as Lord Brown mentioned at [280] a person against whom there is inadmissible intercept evidence that unequivocally demonstrates his guilt, could not on admissible evidence be convicted.

The answer to that difficulty is that compensation in such cases should be derisory only. An award of contemptuously small compensation would show society that the person was in fact guilty, and it would deter such people from seeking compensation.

It should be stressed that this case does not establish a meaning for “miscarriage of justice” beyond its specific statutory context. As Lord Phillips observed at [9], this phrase is capable of having a number of different meanings.

Although there are 284 paragraphs in the case, it is easy to read because the judgments are divided into consistent headings, and the dissenters are left to the end under their own heading. Other topics dealt with are the meaning in this context of the phrase “a new or newly discovered fact” and the relationship between the presumption of innocence and a claim for compensation.

In New Zealand this sort of compensation is governed by Cabinet Guidelines introduced in 1998. Claimants have to be alive at the time of the application, have served all or part of a term of imprisonment, had their convictions quashed on appeal without an order for retrial (or have received a free pardon), and must be able to prove on the balance of probabilities that they are innocent of the crime for which they were convicted. This is a more restrictive entitlement than in the UK. There is, however, a “residual discretion” that allows the Crown to consider claims falling outside the guidelines in “extraordinary circumstances” where it is in the interests of justice to do so. So, where a person is acquitted following a retrial he is outside the guidelines, so he must show “extraordinary circumstances” to demonstrate that it is in the interests of justice to award him compensation. This may well mean that if such a person is able to show to a standard higher than the balance of probabilities that he is in fact innocent, it would be in the interests of justice to award him compensation. The guidelines are not, therefore, unduly concerned with failure to compensate people who are really innocent, and they are arguably over-concerned with avoiding compensating people who are really guilty. Some flexibility in the quanta of awards could be a way of making the scheme more just.

The New Zealand Law Commission had recommended the requirement of proof of innocence and para 127 of its 1998 Report No 49 “Compensating the Wrongly Convicted” received citation in Adams, R (on the application of) v Secretary of State for Justice: [47] Lord Phillips, [173, 175] Lord Kerr. The Commission proceeded on the basis that actual innocence is the justification for compensation, but the majority of the United Kingdom Supreme Court held that excluding from entitlement people who no longer seemed to be guilty but whose innocence could not be established was a heavy price to pay for ensuring that no guilty person ever gets compensation (Lord Phillips at [50]).

The NZLC Report starts with a controversial assertion, saying in para 1 “The essence of a free society is the freedom of a law-abiding citizen to act without interference by the state.” That begs the very question the state has been unable to answer lawfully. In view of the Report’s conclusions, it is a tyrant’s assertion, requiring the citizen to prove he is law-abiding. Instead, the essence of a free society is the freedom of all people, whether law-abiding or not, to act without unlawful interference by the state. The rule of law requires the state to prove that interference with the citizen’s liberty is justified. The United Kingdom Supreme Court’s majority approach survives the moral analysis advocated by Dworkin (see my review of “Justice for Hedgehogs” 25 April 2011), whereas the New Zealand one does not.

To put or not to put?

I suppose the most general lesson that can be learnt from the fact-specific decision of the High Court of Australia in Braysich v R [2011] HCA 14 (11 May 2011) is that rulings on whether there is sufficient evidence, at the conclusion of the case for the defence, for a defence to be put to the jury should not be based on narrow assumptions about how the evidence in the case is to be analysed.

Again, in the most general of terms, where it is an offence to do X but there is a statutory defence of absence of intention to do Y, and where the defendant has denied doing X, evidence about X might also be relevant to Y.

So, where a sharebroker facilitates the transfer of legal ownership of shares, when in fact the beneficial ownership does not change, he is deemed to have committed the offence of creating a false appearance of active trading in those shares, but he has a defence if he proves he did not have the purpose of creating that appearance (see para [5-8] of Braysich for the relevant legislation). The circumstances in which he transferred the shares may be relevant to whether he had the proscribed purpose. When he denies knowing that the beneficial ownership did not change, evidence of his good character will be relevant to that, and it will also be relevant to whether he had the proscribed purpose.

Obvious though this may be in retrospect, the Court split 3-2 on whether the defence case was sufficient to raise the defence of absence of the proscribed purpose. Heydon J agreed with Bell J and added [54] that all the defendant had done at trial was to point to evidence which, although not inconsistent with the absence of purpose did not actually support absence of purpose. Bell J [116] considered that there was no evidence of any purpose other than the impugned purpose. Although the defendant had acted for fees for valued clients, that could not, without more, establish that it was probable that he did not have that purpose. The majority, French CJ, Crennan and Kiefel JJ, held that on the view most favourable to the defendant the jury could well have asked whether it was really likely that an honest man who is acting on instructions from reputable people, who he has no reason to believe have a dishonest purpose, would himself have that dishonest purpose when he is aware of the business rules and the law [47], and that therefore the defence should have been put to the jury.

There was no disagreement about what the law is on whether a defence should be put to the jury. The relevant question in this case was [36]:

“In a case in which both the legal burden and the evidential burden rest upon the accused – is there evidence which, taken at its highest in favour of the accused, could lead a reasonable jury, properly instructed, to conclude on the balance of probabilities that the defence had been established?”

Why the big difference between the majority and the minority? I suspect the minority have taken the stance that unless an individual item of evidence gives rise to a probability of innocence greater than 0.5, it does not support the defence. This approach allows the minority to say that evidence that is merely consistent with absence of purpose does not support absence of purpose. The correct way to look at this, as Bayesian logicians tell us, is to ask what is the probability of getting the evidence, assuming the defendant is innocent, compared to the probability of getting it, assuming that he is guilty. Comparison of those probabilities, assessed intuitively, would indicate that each of the items of evidence relied on by the defendant was more likely to exist if he was innocent than if he was guilty, so they individually and in combination support the defence.