Thursday, February 09, 2006

Rushing to judgment

Cutting to the chase in trials can look unfair. Where legal submissions are necessary, judges may be tempted to deal with them peremptorily, risking the appearance of bias. The bounds of proper judicial conduct were crossed at the trial that was the subject of the appeals jointly heard in Antoun v R [2006] HCA 2 (8 February 2006).

The judges of the High Court of Australia all agreed that the trial had been handled in a way that created the appearance that the judge was biased against the defence. This "apprehended bias", is

"the appearance of a possibility of an absence of an impartial mind on the part of the judge"

(per Callinan J at para 83, with whom the other members of the Court agreed). As Kirby J noted, para 28:

"…The manifest observance of fair procedures is necessary to satisfy the requirements not only of fairness to the accused but also of justice before the public so that they may be satisfied, by attendance or from the record, that the process has followed lines observing basic rules of fairness. Excessively telescoping the procedures in such cases can lead to a sense of disquiet on the part of the accused, and of objective observers whose attitudes, where relevant, must be represented, and given effect, by appellate courts."

In this case the judge, on being told at the close of the prosecution case, that the defence counsel for each accused would make submissions that there was no case to answer, immediately said that such submissions would be rejected. This was said before an opportunity to make those submissions was given. It was a trial by judge alone, but that is not material because even in a jury trial such submissions would be heard in the absence of the jury.

Callinan J held that the judge’s conduct gave an appearance of pre-judgment, and therefore, para 86:

"It follows that the apparent strength of the respondent's case, and the weaknesses of the appellants' defence cannot be used as justification or excuse for the trial judge's expressions of a determination to reject submissions foreshadowed, but not yet made and developed."

Gleeson CJ put it this way, para 23:

"…The judge regarded this as a strong case of extortion. He formed the view, with good reason, that the no case to answer submission was likely to be implausible. Yet he should not have decided to reject it without giving counsel an opportunity to put the argument. In the circumstances, that would not have required much time. The way in which the judge dealt with the no case argument, and later with the question of bail, gave rise to an appearance of lack of impartiality. Strong as the case against the appellants appeared to be, they were entitled to a fair hearing."

This case did not include submissions on whether the proviso could be relied upon to dismiss the appeal. However, Kirby J noted, para 49, that the denial of the opportunity to make submissions before decision would probably be a sufficient miscarriage of justice to make the proviso inapplicable, and that, had the case been argued on the proviso, it would have required consideration of Weiss v The Queen [2005] HCA 81 at [45] (see these blogs, 16 January 2006).
We might, by now, anticipate that a case of apprehended bias would be a substantial miscarriage of justice, given the fundamental importance of the appearance of justice, but one must acknowledge that there is, at least, an argument that whereas actual bias would be a substantial miscarriage of justice, apprehended bias falls short of that in cases where there was no loss of a real chance of acquittal.

Consideration of Antoun raises a number of fundamental questions. Is there a difference between so-called "apprehended" bias and "actual" bias? If there is such a thing as apprehended bias, does it fall within the scope of the accused’s right to a fair hearing, or is it an associated right? Current jurisprudence suggests that if it is the former, it is absolute, but if it is only an associated right then it is subject to balancing against competing values. Doesn’t the actual decision in Antoun (that apprehended bias requires a retrial even where the judge’s ruling was correct) mean that the real issue in the case was not whether the hearing had involved bias (whether actual or apprehended), but rather whether there had been a "hearing" on the relevant issue? On analysis, the real reason a retrial was necessary in Antoun is that the accused had been denied the right to a hearing on the issue of whether there had been a case to answer.

Friday, January 27, 2006

Fingered!

What makes fingerprint evidence interesting is the increasing unlikelihood that we all have different fingerprints. This, at least, is what ordinary old you and I might think. Fingerprints are not genetically determined: identical twins have different fingerprints. While their design falls broadly into a few types, at a finer level of detail they seem to be formed randomly.

There are more human fingers than there ever have been. Often, so-called identifications are made from incomplete prints left at the scene of a crime. Part of the print of one finger, it is said, may be sufficient for an expert to say that it is from the accused. However many fingers there are, there are many more "parts of fingers". Yet they are all, so it is said, sufficient to mark us out individually. This compression of information, in a small area of skin, is truly remarkable. Mathematicians can assert that there are something like 10 to the power of 20 different possible patterns in a small area of finger, so, if we accept that, our suspicions about the uniqueness of fingerprints are needlessly alarmist.

Let us assume, however, that the experts are not complacent and that they search with sustained vigour for people who have identical "parts of fingerprints". This amounts to assuming that the experts, who have made a career in reliance on the theory that we all have fingerprints that are unique, are prepared to risk making a discovery that will undermine their livelihoods. I exaggerate, of course: even if many people on earth had the same fingerprints, the finding of one at a crime scene that was, after allowance was made for the distorting effects that occur in real life, not significantly different from the accused's, would, on Bayesean analysis, be highly probative.

Far from being suspicious, the courts are apparently becoming more relaxed about fingerprint evidence. There is judicial acceptance of a trend away from the basing of fingerprint identifications on the number of specified points of identity as between the accused’s print and the sample found at the crime scene. Instead of showing the court enlarged photographs with arrows specifying points on the prints that the court should accept as demonstrating identity, the fingerprint experts are now permitted to base their opinions on the "quality and quantity of information in the images": R v Carter 19/12/05, CA155/05, a phrase used in a technical paper quoted by the Court at para 75.

Important, in this approach, is peer review of the expert witness’s opinion. This review must, in England and Wales, be carried out by three "independent" examiners.

In Carter the Court did not go so far as to say that three independent examiners are needed as peer reviewers before evidence of fingerprint identity will be admissible in New Zealand. Plainly, there would be difficulties finding people who are "independent", let alone people who are sufficiently experienced to qualify as experts, in a pool of talent that is, presumably, rather small. Nevertheless, the Court did reiterate a requirement that Tipping J had made of expert witnesses in the well-known case of R v Calder 12/4/95, HC Christchurch T154/94 (where the accused was ultimately acquitted of poisoning), that the evidence needed to have a "minimum threshold of reliability" to be admissible.

So, the need for the judge to warn the jury to exercise caution before accepting evidence of fingerprint identification where there are relatively few points of comparison (R v Buisson [1990] 2 NZLR 542 (CA)) has apparently been replaced by a more relaxed approach. In Carter, the expert was unwilling or unable to indicate whether there were any points of comparison identified (para 60), but the Court of Appeal held that no particular direction to the jury was needed on this, because the defence, in its closing, had not suggested to the jury that the expert was wrong. Instead, the defence, when the time came to close its case, had merely suggested that there were innocent explanations for the accused’s fingerprint being at the scene.

Incidentally, the real question on which the expert's evidence is relevant, is whether there are any significant differences between the print found at the scene and the accused's fingerprint. As a prosecution witness, the expert will be expected to say there are none, and it is his explanations for apparent differences that are material.

One might wonder whether, in a case where defence counsel had apparently been discouraged (the judge having ruled the expert’s evidence admissible) from a full assault on the credibility of the expert, and where peer review witnesses were not called, the jury were given a proper opportunity to reject the fingerprint evidence.

This raises the question of when a decision made as a matter of trial tactics can be regarded as an error that caused a loss of a real chance of acquittal (assuming that if there was a doubt about the fingerprint identification then there would have been a doubt about guilt); as the Supreme Court held in Sungsuwan (blogged here, 26 August 2005), it is the effect of the error, not its magnitude, that matters.

Update: for a critique of fingerprint evidence, see Gary Edmond, "Fingerprint Evidence in New Zealand's Courts: the Oversight of Overstatement" (2020) 29 NZULR 1. And further: Gary Edmond, Fingerprint Evidence in New Zealand's Courts: the Limits of Legal "Challenge" (2020) 29 NZULR 347.

Tuesday, January 24, 2006

Index to these blog entries

To access my Index to all the entries on this blogsite, up to the end of the last month, click on the link on the right.

Recklessness

What does "reckless" mean? Where a person can be guilty of a crime if he has acted with a state of mind called "recklessness", we need to know whether recklessness is to be given a subjective or an objective meaning.

As all criminal lawyers know, the meaning of recklessness was a matter of huge debate in the latter part of the 20th century. This occurred first in relation to the crime of rape. It was clear that having sexual intercourse intending to do so knowing that the woman did not consent was sufficient for guilt, because "intention" is the usual mental requirement for liability. But what about intending to have sexual intercourse, knowing that there was a risk that the woman was not consenting, but hoping that she was? This state of mind, the deliberate taking of a known risk, is recklessness with a subjective meaning, if the taking of the risk was unreasonable. It is "subjective" because the man knew of the risk.

An illustration of the objective meaning of recklessness is where a man has sexual intercourse, thinking that the woman is consenting, but in circumstances where any reasonable person would have been aware of the risk that she was not. This is "objective" because it refers to what the reasonable person would have known, not what the man himself knew.

Although it was not done with great clarity, it seems that the subjective meaning of recklessness was applied by the House of Lords in R v Morgan [1976] AC 182, where the offence in question was rape. It was felt, however, that policy required an objective interpretation of recklessness for some offences, such as arson: R v Caldwell [1982] AC 341 (HL). This view was changed, and Caldwell was overruled, in R v G [2004] 1 AC 1034 (HL). Lord Bingham put it this way, in para 32:

"conviction of serious crime should depend on proof not simply that the defendant caused (by act or omission) an injurious result to another but that his state of mind when so acting was culpable… . The most obviously culpable state of mind is no doubt an intention to cause the injurious result, but knowing disregard of an appreciated and unacceptable risk of causing an injurious result or a deliberate closing of the mind to such risk would be readily accepted as culpable also. It is clearly blameworthy to take an obvious and significant risk of causing injury to another. But it is not clearly blameworthy to do something involving a risk of injury to another if … one genuinely does not perceive the risk. Such a person may fairly be accused of stupidity or lack of imagination, but neither of those failings should expose him to conviction of serious crime or the risk of punishment."

In New Zealand, the subjective meaning of recklessness is applied: R v Harney [1987] 2 NZLR 576 (CA), R v Tihi [1989] 2 NZLR 29 (CA).

This whole topic was aired recently in the High Court of Australia, considering New South Wales legislation in Banditt v R [2005] HCA 80 (15 December 2005). This was a case of rape, and the majority judges (Gummow, Hayne and Heydon JJ) held that the subjective meaning of recklessness applied: it was necessary, for the accused to be guilty, that he be proved to have been aware of the risk that the complainant was not consenting. Callinan J, although agreeing in the dismissal of the appeal, thought that the legal wrangling over the meaning of recklessness was so complex that the legislature must have intended simply that the ordinary meaning, as determined by the jury, should apply. This, with respect, is plainly wrong, for the ordinary meaning of recklessness can include (COD) "disregarding the consequences or danger, etc; lacking caution; rash." When one remembers that sexual intercourse is often indulged in rashly, it is plain that this approach would cast the net too far.

Monday, January 23, 2006

Illegal acquittals

Objections to the power proposed to be given to the prosecution, in some circumstances, to appeal against acquittals, are usually advanced on the grounds that the finality of an acquittal is a fundamental common law right, a matter of "due process".

The Privy Council has considered this in The State v Boyce (Trinidad and Tobago) [2006] UKPC 1 (17 January 2006). The judgment notes different usages of the term "due process": there is a broad sense, (para 13):

"In this sense, the concept of due process incorporates observance of all the mandatory requirements of criminal procedure, whatever they may be."

And there is a narrow, constitutional, sense (para 14):

"… those fundamental principles which are necessary for a fair system of justice."

The absolute prohibition at common law of a challenge to the finality of an acquittal was not, the Judicial Committee held, part of the narrower meaning of due process (para 15). This conclusion required classifying the finality of an acquittal as a "broad principle":

"…the broad principle that a person who has been finally convicted or acquitted in proceedings which have run their course should not be liable to be tried again for the same offence is a fundamental principle of fairness. It is recognised as such in many constitutions (see, for example, section 20(8) of the Constitution of Jamaica) and in international human rights instruments: see for example article 14.7 of the UN International Covenant on Civil and Political Rights and article 4 of Protocol No 7 to the European Convention for the Protection of Human Rights and Fundamental Freedoms. But [their Lordships] do not think that the principle is entirely without exceptions (see, for example, article 4.2 of Protocol No 7) and they certainly do not think that it is infringed by the prosecution having the right to appeal against an acquittal. The possibility of such an appeal is accommodated in the qualification of the principle ("save upon the order of a superior court made in the course of appeal proceedings relating to the conviction or acquittal") in all the Caribbean constitutions to which their Lordships were referred (Jamaica, Barbados, The Bahamas, Grenada, Dominica, Saint Lucia, Saint Vincent and the Grenadines, Guyana, Antigua and Barbuda, Belize, Anguilla, St Christopher and Nevis, Turks and Caicos Islands, Monserrat) as well as the international instruments which their Lordships have mentioned."

Thus, the proceedings may not have "run their course" until any right that the prosecution may have to appeal against an acquittal has been exercised. Once it has, the broad principle of finality would apply.

It is sensible, of course, to permit an appellate court to correct errors that have occurred in the application of the law during a trial, whichever side they may have favoured. This point was made in para 16:

"…There is nothing particularly unfair or unjust about a statutory rule which enables an appellate court to correct an error of law by which an accused person was wrongly discharged or acquitted and order that the question of his guilt or innocence be properly determined according to law. Such a rule exists in many countries. It is true that in Davern v Messel (1984) 155 CLR 21 the High Court of Australia said that the common law rule was so ancient and well settled that a statute giving a right of appeal in general terms against the decisions of a court would not be construed as allowing the prosecution to appeal against a decision in favour of an accused person. Express language was needed. But the court did not suggest that the absence of a right of appeal formed part of fundamental due process and in the present case the language of section 65E [of the Supreme Court of Judicature Act] is clear enough."

Other interesting points made in this case are:
  • Proceedings on indictment start with the filing of an indictment; where an information is laid indictably, that is just a preliminary to indictable proceedings but does not form part of them (para 22).
  • An "error of law" can arise wherever the judge had made a wrong determination, regardless of whether it concerns pure law, or a mixture of fact and law. It thus can include rulings on admissibility of evidence and on whether there is a case to answer (para 24).
  • In deciding whether a witness qualifies as an expert, it is wrong (and this was one of the errors made by the judge in this case) to focus only on the witness’s paper qualifications. The judge was also wrong to call another witness solely to comment on whether another witness should qualify as an expert (para 25, 26).
  • Where an error of law has occurred, and an acquittal has resulted, the question whether a new trial should be ordered can depend on whether a fair trial is still possible. Here, the error of law was corrected (ie the provision creating the right to challenge an acquittal was held to be constitutional), but the appeal was dismissed because it could not be said that a new trial would be fair, 9 years having elapsed since the events in question (para 26, 27).

Friday, January 20, 2006

Three major themes

Three topics of general interest were touched upon by the Privy Council in Grant v R (Jamaica) [2006] UKPC 2 (17 January 2006): (1) the interrelationship between the right to a fair trial, and the balancing of constituent rights; (2) the discretion to exclude evidence where its prejudicial effect exceeds its probative value, and the relationship of this discretion to the right of the accused to a fair trial; and (3) whether the proviso can be applied to dismiss an appeal when potentially significant evidence was never before the jury (including, on the facts of this case, an illustration of what is "potentially significant" evidence).

Right to a fair trial, and balancing

The Privy Council referred to laws similar to those in effect in Jamaica, including decisions of the European Court of Human Rights. The established model is that the accused’s right to a fair trial is absolute, but rights that are subsidiary to that are subject to balancing against each other. See Mathias, "The Accused’s Right to a Fair Trial: Absolute or Limitable?" [2005] New Zealand Law Review 217.

At para 17 of Grant, the Privy Council put it this way:

"The Strasbourg court has time and again insisted that the admissibility of evidence is governed by national law and that its sole concern is to assess the overall fairness of the criminal proceedings in question: see, for example, Kostovski v Netherlands (1989) 12 EHRR 434, para 39; Windisch v Austria (1990) 13 EHRR 281, para 25; Lüdi v Switzerland (1992) 15 EHRR 173, para 43; Saidi v France (1993) 17 EHRR 251, para 43; Doorson v Netherlands (1996) 22 EHRR 330, para 67; PS v Germany (2001) 36 EHRR 1139, para 19. The specific rights set out in article 6(3) of the European Convention (and thus, by analogy, section 20(6) of the [Jamaican] Constitution [see below]) are "specific aspects of the right to a fair trial" (Kostovski v Netherlands, above, para 39) or "particular aspects of the right to a fair trial" (Doorson v Netherlands, above, para 66), and the right to a fair trial can never be compromised in any circumstances. But the constituent rights in article 6 and section 20(6) are not themselves absolute: Brown v Stott (Procurator Fiscal, Dunfermline) [2003] 1 AC 681, 704. The Strasbourg court has been astute to avoid treating the specific rights set out in article 6 as laying down rules from which no derogation or deviation is possible in any circumstances. What matters is the fairness of the proceedings as a whole."

Section 20(6) provides:

"(6) Every person who is charged with a criminal offence—
(a) shall be informed as soon as reasonably practicable, in a language which he understands, of the nature of the offence charged;
(b) shall be given adequate time and facilities for the preparation of his defence;
(c) shall be permitted to defend himself in person or by a legal representative of his own choice;
(d) shall be afforded facilities to examine in person or by his legal representative the witnesses called by the prosecution before any court and to obtain the attendance of witnesses, subject to the payment of their reasonable expenses, and carry out the examination of such witnesses to testify on his behalf before the court on the same conditions as those applying to witnesses called by the prosecution; and
(e) shall be permitted to have without payment the assistance of an interpreter if he cannot understand the English language."


These, then, are capable of being compromised in the balancing exercise, which the Privy Council described in Grant (also in para 17) as follows:

"Just as section 13 of the Constitution recognises that individual rights cannot be enjoyed without regard to the rights of others, so the Strasbourg court has recognised the need for a fair balance between the general interest of the community and the personal rights of the individual, and has described the search for that balance as inherent in the whole Convention: Sporrong and Lönnroth v Sweden (1982) 5 EHRR 35, para 69; Sheffield and Horsham v United Kingdom (1998) 27 EHRR 163, para 52; Brown v Stott, above, p 704. Thus the rights of the individual must be safeguarded, but the interests of the community and the victims of crime must also be respected. An example, not based on the present facts, illustrates the point. In Jamaica, as in England and Wales, as already noted, the statement of a witness may be adduced in evidence if he is shown to have absented himself through fear of the consequences to him if he gives evidence. In the case of a prosecution witness, such fear is likely to have been induced by or on behalf of a defendant wishing to prevent adverse evidence being given. As observed by Potter LJ in R v M(KJ) [2003] EWCA Crim 357, [2003] 2 Cr App R 322, para 59, echoed by Waller LJ in R v Sellick [2005] EWCA Crim 651, [2005] 1 WLR 3257, paras 36, 52-53, it would be intolerable if a defendant shown to have acted in such a way could rely on his human rights under article 6 (or section 20) to prevent the admission of hearsay evidence. Where a witness is unavailable to give evidence in person because he is dead, or too ill to attend, or abroad, or cannot be traced, the argument for admitting hearsay evidence is less irresistible, but there may still be a compelling argument for admitting it, provided always that its admission does not place the defendant at an unfair disadvantage."

Obviously, where the need to ensure fairness to the accused is paramount, as clear a definition, as is possible, of fairness needs to be established. It is on this central point that further elaboration is needed. I have suggested, in the article referred to above, that a fair trial meets two essentials: (1) there is an opportunity for the jury properly to apply the law, and (2) that application of the law is to the facts determined without bias.

The concern in Grant was whether the admission of the hearsay evidence would create bias against the accused by undermining his ability to challenge that evidence; whether, in other words, admission of the hearsay statement created an inequality between the prosecution and the defence.

The Board carefully set out the safeguards that the law provided for accused persons faced with such hearsay evidence (para 21). These included strict criteria for establishing the unavailability of the witness as a precondition for admission of his written statement, enhanced rights of the defence to call evidence to challenge the truthfulness of the evidence, and a need for the judge to give special directions to the jury cautioning them on what weight to give the hearsay evidence. Also among these safeguards is the next topic to be considered here: the discretion to exclude the evidence where its prejudicial effect exceeds its probative value.

Probative value and prejudicial effect

I have recently written on this topic: "Probative value, illegitimate prejudice and the accused’s right to a fair trial" (2005) 29 Crim LJ 8. In essence, my point is that the discretion to reject evidence on this ground, which involves a so-called balancing of probative value against prejudicial effect, is flawed because the balancing, or weighing, notion suggests that only a high level of unfairness to the accused will outweigh highly probative evidence. That in turn suggests that trial fairness doesn’t matter so much if the accused is obviously guilty. The real question, I suggest, is whether the evidence has sufficient probative value to be admissible bearing in mind the law’s reluctance to allow hearsay evidence; if it does have sufficient probative value, then it is admissible, subject to the trial still being fair for the accused.

The Privy Council has taken this approach in Grant, para 21:

"Section 31L acknowledges the discretion of the court to exclude evidence if it judges that the prejudicial effect of the evidence outweighs its probative value. In R v Sang [1980] AC 402, some members of the House of Lords (notably Lord Diplock at pp 434, 437 and Viscount Dilhorne (pp 441-442)) interpreted this discretion narrowly, and in Scott v The Queen [1989] AC 1242, 1256-1257, the Board appears to have accepted that reading. It is not, however, clear that the majority in R v Sang favoured a similarly narrow interpretation (see Lord Salmon at pp 444-445, Lord Fraser of Tullybelton at p 449 and Lord Scarman at pp 453, 454, 457). In any event, it is, in the opinion of the Board, clear that the judge presiding at a criminal trial has an overriding discretion to exclude evidence which is judged to be unfair to the defendant in the sense that it will put him at an unfair disadvantage or deprive him unfairly of the ability to defend himself. Such a discretion has been recognised by the Court of Appeal in R v Donald White (1975) 24 WIR 305, 309, and R v Michael Barrett, above. It has been recognised by the Board in Scott v The Queen, above, pp 1258-1259 and Henriques v The Queen [1991] 1 WLR 242, 247: both these appeals concerned the admission of depositions, but the need for a judicial discretion to exclude is even greater when the evidence in question has never been given on oath at all. In England and Wales, the discretion has been given statutory force: see section 25(1) of the Criminal Justice Act 1988; R v Lockley [1995] 2 Cr App R 554, 559-560; R v Gokal [1997] 2 Cr App R 266, 273; R v Arnold [2004] EWCA Crim 1293, para 30. Conscientiously exercised, this discretion affords the defendant an important safeguard."

Thus, whether one takes a broad or a narrow approach to the so-called balancing, or weighing, of probative value against prejudicial effect, doesn’t matter, because the overriding criterion is whether admission of the evidence would make the trial unfair for the accused.

In para 26 of Grant the Board analysed the circumstances of the trial and gave reasons for their conclusion that the admission of only one of two hearsay statements caused unfairness to the accused. Grant was accused of murder. He had been approached by the victim in a threatening manner while urinating by some bushes at night. He claimed the person had a gun. He took out his own gun and fired at the victim several times. The victim turned and ran. Grant chased him and fired several shots again. The evidence diverged crucially at this point: the hearsay statement that was admitted in evidence was of a witness, Bryant, who said he saw the victim lying face down on the ground when he was approached and shot several times. The victim died from wounds he had received in the back. Grant denied approaching the victim and shooting him like that, saying in evidence that he fired after him because he thought he was still at risk of attack: he relied on self defence. A hearsay statement by another person, Kinglock, was excluded from evidence. It would have confirmed the first part of the episode: someone approaching Grant while he was urinating, then shots being heard and the person running and throwing something (this, if Grant’s evidence had credibility, would have been a gun) away.

In summary, the trial was unfair because (para 26):

"It is, in the Board's opinion, plain that fairness required the admission of Kinglock's statement. If admitted, it might not have been understood to exonerate the appellant. The proliferation of shots to the back of the deceased was a formidable problem for him to overcome. The jury might have convicted anyway, and been entitled to do so. But the jury should have known how, according to Kinglock, in large part corroborating the appellant, the incident began. The appellant was entitled to have his case assessed, and his own evidence evaluated, in the light of all the available evidence. … [the judge] could have invited prosecuting counsel to adduce Kinglock's statement in evidence. Had that invitation, improbably, been declined, the judge could, on grounds of fairness, have declined to admit Bryant's statement unless Kinglock's statement were also admitted or could, in the last resort, have introduced the statement of Kinglock herself (R v Oliva [1965] 1 WLR 1028, 1035-1036). … The Crown having given the defence notice of its intention to adduce the statement of Kinglock as well as the statement of Bryant at the trial, defence counsel appears to have been taken by surprise when on the second day of the trial prosecuting counsel applied to adduce the statement of Bryant, but not that of Kinglock. … It was, however, the responsibility of prosecuting counsel and the trial judge to ensure that the proceedings were fair, and they failed to do so. This failure was compounded by an inadequate direction on Bryant's evidence. The jury were given no encouragement to scrutinise it with particular care, and were not alerted to apparent discrepancies between it and the evidence of Constable Wynter (or, of course, the statement of Kinglock)."

These points illustrate the two aspects of trial fairness: the need to avoid bias in the opportunity the jury has to determine the facts, and the need to ensure the jury has an opportunity correctly to apply the law.

The proviso

Finally, Grant illustrates consideration of whether to apply the proviso. That is, whether to dismiss the appeal against conviction notwithstanding that errors at trial had occurred. The statutory criterion is whether errors had amounted to a "substantial miscarriage of justice". The meaning of this phrase is critical, and it is still being worked out after over a century of consideration by courts in the common law countries.

As is usual, the proviso did not receive detailed treatment by the Board, which simply concluded, para 27:

"It would not be appropriate to apply the proviso in a case where potentially significant evidence was never before the jury."

This means that, in the context of Grant, the evidence of Kinglock was "potentially significant", even though "it might not have been understood to exonerate the appellant" (para 26, above). It was corroboration of Grant’s version of how the incident started that was "potentially significant": "The appellant was entitled to have his case assessed, and his own evidence evaluated, in the light of all the available evidence" (ibid). This was so, even though the partial defence of provocation was not relied on. The second part of the incident, not covered in Kinglock’s evidence, would be most unlikely to be the basis for a successful defence of self defence, as the Board recognised: "The proliferation of shots to the back of the deceased was a formidable problem for him to overcome. The jury might have convicted anyway, and been entitled to do so" (ibid).

Plainly, the rationale for the application of the proviso, not elaborated by the Board, is that the trial was not fair. No matter how strong the case, an unfair trial cannot sustain a conviction.

Monday, January 16, 2006

Revising history

Where a trial has, in the opinion of an appellate court, involved a miscarriage of justice, the appeal against conviction should be allowed,

"Provided that the Court of Appeal may, notwithstanding that it is of opinion that the point raised in the appeal might be decided in favour of the appellant, dismiss the appeal if it considers that no substantial miscarriage of justice has actually occurred."

These statutory words are copied in many jurisdictions throughout the common law world. The High Court of Australia has recently given its attention to their meaning in Weiss v R [2005] HCA 81 (15 December 2005).

One of the fundamental problems faced by those who have to interpret texts of any real vintage (this one is early 20th century), is whether to try to determine what those who used the language intended it to mean then, and to apply that meaning now, or, whether to give the text the meaning that best serves the interests of society now.

The High Court of Australia in Weiss used the former approach, on the rather spurious basis that this was necessary as a way of rejecting interpretations that would, in effect, result in all appeals being allowed wherever there was any error (miscarriage of justice) at trial, and interpretations that are based on a perceived need to protect a "right" to trial by jury by avoiding the appellate court’s substitution of its own verdict for that of the jury at a new trial. These reasons for the Court's approach are spurious because the currently accepted approach to the proviso (albeit not fully worked out) does not require retrials merely because error is found to have occurred at the original trial: the requirement is that the error affected the outcome of the trial.

Having gone that far with history, using it as justification for attacking a misperceived fault, the High Court then changed tack, emphasising (para 31) that it is the language of the statute, not secondary sources or materials that matters. However, the Court had, by its view of history, set itself on a new interpretative course. But this cannot disguise the real question, namely, which approach to determining the meaning of the statutory language is appropriate, historical meaning or current policy?

Three fundamental propositions were stated, para 39:

"Three fundamental propositions must not be obscured. First, the appellate court must itself decide whether a substantial miscarriage of justice has actually occurred. Secondly, the task of the appellate court is an objective task not materially different from other appellate tasks. It is to be performed with whatever are the advantages and disadvantages of deciding an appeal on the record of the trial; it is not an exercise in speculation or prediction. Thirdly, the standard of proof of criminal guilt is beyond reasonable doubt."

These, it might be thought, are not particularly informative. They are intended to be borne in mind by appellate courts when they apply what may now be called the Weiss approach to the proviso: the court must make its own assessment of the whole of the record of the trial to decide whether it proved the accused’s guilt beyond reasonable doubt (para 41). The proviso can only be applied (ie it is a necessary, but not always a sufficient condition – as indicated next - for application of the proviso) if the court (para 44)

"… is persuaded that the evidence properly admitted at trial proved, beyond reasonable doubt, the accused's guilt …"

However, the proviso should not be applied in some cases (para 45):

"… there may be cases where it would be proper to allow the appeal and order a new trial, even though the appellate court was persuaded to the requisite degree of the appellant's guilt. Cases where there has been a significant denial of procedural fairness at trial may provide examples of cases of that kind."

Pursuant to this approach, the appellate court acts as a substitute jury, albeit one that has not seen or heard the witnesses. Compensating for that absence of contact with the witnesses, the appellate court is apparently to be particularly mindful of the high standard of proof.

Well, how did the High Court deal with the case before it in this appeal?

The issue, on this approach, came down to whether a reasonable doubt about the accused’s guilt could have existed in view of a contest between whether a confession by him in his (second) police interview was made as a result of improper inducement by a police officer. The miscarriage of justice at trial had been wrongful admission of evidence of the accused’s bad character. That is, could the accused’s explanation of why he made the confession have been rejected because of the miscarriage of justice?

Put this way, it would probably appear to most people that there should be a re-trial. However, the High Court remitted the case to the Court of Appeal (Victoria) so that it could carefully examine the record of the trial in the light of the approach to the proviso it laid down (para 57):

"… one question for the Court of Appeal was whether, considering all of the evidence at trial, these matters of character could be put aside as unimportant side issues when viewed in the context of the whole trial, particularly as the evidence in the trial included the powerful testimony of confessions to police which the appellant did not contest making, although he sought to explain how they came about. If they could, attention could focus upon whether the videotaped confession (which the appellant had undoubtedly made) established, beyond reasonable doubt, his guilt of murder. Or was there a reasonable possibility that he had made a false confession?"

With respect, it is difficult to imagine how evidence suggesting the accused has little appreciation of the difference between right and wrong can possibly be ignored in a contest of credibility between the police officer and the accused.

The cumbersome Weiss approach to the proviso may lead a court to overlook the obvious. It is an unnecessary rejection of a century of judicial progress, without a demonstration of any particular problem, let alone a search for a less drastic remedy.

If an appellate court does not apply the proviso, it may order a new trial. It therefore seems strange, given this ability to return the case to a jury, for Weiss to require the court to ignore what effect an error might have had (or, the absence of an error might have) on the jury - the current approaches - and instead focus (if "focus" is the word for such an obscure exercise) on the court's perception, taking into account the guilty verdict, of the likelihood of guilt.

Furthermore, Weiss, perhaps unintentionally, emphasises the difference between the broad approach taken to matters that come before appellate courts pursuant to the royal prerogative (Mallard v R [2005] HCA 68 (15 November 2005), blogged here 24.11.05, not cited in Weiss), and the narrower approach to those that come pursuant to the ordinary criminal appeals procedure. Under both Mallard and Weiss, the court must consider all of the evidence. But under the Mallard (prerogative) approach, the court is not restrained by the view of the facts that must have been taken by the trial jury. One might wonder why, on an ordinary criminal appeal, as held in Weiss, the jury's view should be a constraint. Whether it is good policy to maintain such a distinction is doubtful. Republican Australians might be surprised to learn that they get better quality justice via the Queen.

Thursday, December 22, 2005

The more, the merrier

How straight-laced are Quebeckers? One would have thought this was a silly-season question. And yes, the people of Quebec are called Quebeckers. In R v Labaye [2005] SCC 80 (21 December 2005) the Supreme Court of Canada had to decide whether acts of indecency included group sex by consenting adults in an environment where any spectators were also voluntary on-lookers. The larger a court, the more likely it is that there will be dissent. Here, the court split 7 – 2.

The majority held that in the circumstances the group sex was not indecent. What is remarkable about this is the complexity of the legal test for indecency, which was summarised at para 62 as:

"Indecent criminal conduct will be established where the Crown proves beyond a reasonable doubt the following two requirements:

1. That, by its nature, the conduct at issue causes harm or presents a significant risk of harm to individuals or society in a way that undermines or threatens to undermine a value reflected in and thus formally endorsed through the Constitution or similar fundamental laws by, for example:
(a) confronting members of the public with conduct that significantly interferes with their autonomy and liberty; or
(b) predisposing others to anti-social behaviour; or
(c) physically or psychologically harming persons involved in the conduct, and


2. That the harm or risk of harm is of a degree that is incompatible with the proper functioning of society."

The vetting of members of this private club (L’Orage Club, Montreal) meant, held the majority, that there was no such harm as was required by this test.

The minority placed less emphasis on the private nature of the club, and applied what had hitherto, in their view, been the traditional test, the community standards. Interestingly, they referred to JS Mill (para 105) as underpinning the majority’s emphasis on the need for harm, and they criticized this on the grounds that behaviour may be criminalised without there being an identifiable harm:

"In the case at bar, the offence relates to social morality. To place excessive emphasis on the criterion of harm will therefore make it impossible to give effect to the moral principles in respect of which there is a consensus in the community."

Noting that evidence indicated that only 2 to 5% of Quebeckers engaged in group sex (para 138), the minority concluded (para 153):

"Our analysis does not permit us to conclude that the Canadian community would tolerate the performance, in a commercial establishment to which the public has easy access, of group sexual activities on the scale of those that took place in this case."

The Canadian Charter of Rights and freedoms was not cited in this case, but it makes the rights and freedoms it guarantees "subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society" (clause 1). Relevant Charter rights are (clause 2) freedom of conscience, thought, belief, opinion, peaceable assembly, and association. (And see the comparable provisions of NZBORA, s 5, 13, 14, 16, 17.) Even if the minority in Labaye is correct, and the Canadian community would not tolerate group sex in the circumstances in which it occurred in that case, the further question should be whether that intolerance is justified in a free and democratic society. The harm-based analysis applied by the majority surely has a role here.

A "companion case" was also decided yesterday by the same Court, with the same split in judges: R v Kouri [2005] SCC 81 (21 December 2005). Here, the question was whether group sex in a bar was indecency, where entrance was restricted to people who agreed that they were "liberated". Incongruously, entrance was also restricted to "couples", a rather charming nod in the direction of propriety. Important for the majority was the effective restriction of entry to those who knew very well what they were in for.

I should add a technical point. In Kouri, the minority judges dissented by using the same reasoning they employed as the minority in Labaye. Strictly speaking, the law was established in Labaye, and the minority were compelled to follow it. If they were to reach a conclusion different from that of the other Judges, they should have done so by applying lawful reasoning. For example, they may have decided that there was sufficient social harm arising from the relevant conduct to make it criminally indecent. Alternatively, and admittedly, impractically, the minority would should have required argument on whether the law as established in Labaye should be changed.

Monday, December 19, 2005

The evil that men think

If I am offended, not by what I see a person doing, but by what I think he might do later, have I observed behaviour that the criminal law calls "offensive"?

Further, if I have gone out of my way to see what the person is doing, can I complain? He might have been unobservable by all but the most inquisitive.

Obviously, judges may disagree over what amounts to offensive behaviour. They may also disagree over whether offensive behaviour is nevertheless insufficiently serious to warrant the intervention of the criminal law.

Cases on questions such as these are encountered by all first-year law students. Far from being confined to serve as minor intellectual irritants for novices, such matters also cause disagreement among senior judges.

In R v R (2005) 21 CRNZ 610 (CA) the issues included whether it was offensive behaviour for the appellant to have concealed himself in his curtained van and taken pictures of uniformed schoolgirls as they arrived at school. He could only be observed with difficulty, and was observed by a police officer who went up to the van and peered through a gap in the curtains. One might wonder, at this point, whether such conduct by the police officer might itself be offensive. In any event, one of the Judges in the Court of Appeal, who is not named in the Court’s single judgment, thought the appellant’s behaviour was not, in law, "offensive". The conduct was, this Judge thought, amenable to innocent interpretation (para 39):

"Had the photographer been, for example, a police officer taking photographs for the purposes of a criminal investigation or a paparazzi [sic] taking photographs for a newspaper or magazine, the observable behaviour would have been identical."

But would it necessarily be sufficient to know more? If there was evidence, for example, that the photographs were to be used for sexual gratification, would the taking of the photographs then be offensive? This point did not require consideration by the majority in R, because they found that the behaviour was, albeit marginally, offensive.

This point was, however, considered in Rowe v Police 12/12/05, John Hansen J, HC Dunedin CRI 2005-412-000051. The same Mr "R" took photographs in the University Library. Perhaps learning from his earlier brush with the law, he did not, on this occasion, resort to subterfuge. While his actions were in plain view, there was insufficient evidence in the case to prove what he was photographing. On the point of whether the behaviour would be offensive if more was known, the Judge observed (para 46):

"…It is the action that must be offensive. As a matter of logic, if the observer has no knowledge of what is being photographed the content of the photographs cannot be used to create the necessary level of offence. But the Judge has gone further. She has referred to the downloading of material to the laptop, and possibly the other circumstances as well, as creating in the mind "the need to determine legitimacy". It is unclear from [this] how the need to determine legitimacy would arouse feelings of anger and/or potential disgust [the legal test for offensiveness], or what is the evidential basis for such a finding. While it may raise a need to investigate further, it falls short of what is required in the criminal sense."

His Honour also noted that in the earlier case, the Court of Appeal had referred to the possibility that evidence of the defendant’s purpose "may be relevant in considering and weighing to strike the balance between competing interests" (para 47). That, however, was said in relation to purpose that is evident to those observing the behaviour (see paras 31-34 of R v R, above).

Unfortunately, the Supreme Court refused leave to appeal from the decision in R v R, because the question of law was not framed in sufficiently general terms to come within the Court’s jurisdiction: Rowe v R [2005] NZSC 40 (23 June 2005).

This discussion highlights the importance of the distinction between the impact on us of our own thoughts, and the impact on us of other people’s conduct.

Friday, December 16, 2005

Proof or believability

One of the difficult distinctions in the law of evidence concerns the use of statements made by a witness on a previous occasion to contradict what the witness now says in court. On the one hand, such statements impeach credibility: the court may place less reliance on what the witness now says, because the witness has said something different before. In this situation, reducing reliance on what the witness now says does not amount to positive proof to the contrary. On the other hand, such statements could be treated as proof of what they say, and the court might conclude that the facts were as the witness had previously said, not as the witness now says.

The Supreme Court of Canada, in R v Henry [2005] SCC 76 (15 December 2005), has held that an accused’s prior testimony at a miscarried trial may be used, to contradict evidence he gives at a subsequent trial, as proof of the truth of what was previously said. At para 48 it was put this way:

" …If the contradiction reasonably gives rise to an inference of guilt, s. 13 of the Charter [the right not to be compelled to give evidence; comparable to s 25(d) New Zealand Bill of Rights Act 1990] does not preclude the trier of fact from drawing the common sense inference."

The different approaches were summarised in para 8:

" … both parties view with scepticism the idea that the trier of fact can truly isolate the purpose of impeaching credibility from the purpose of incrimination. They agree on the problem but disagree about the solution. The appellants’ solution … is that unless the statements used to contradict the present testimony were innocuous when made at the first trial, and still innocuous at the second trial, they should be altogether excluded, i.e. even for the limited purpose of challenging credibility. … Otherwise, they fear, the contradictions may well be used by the trier of fact for the forbidden purpose of incrimination. The Crown also recognizes the troublesome nature of the distinction but … says that fair trial considerations absolutely require that the contradictions in the evidence of an accused be exposed. The Crown then goes further … in saying that the trier of fact should be able to make of the contradictions what it wishes, including drawing an inference of guilt, and indeed that a realistic appraisal of the trial process permits no other conclusion, human nature being what it is."

The resolution of this problem is reasonably easy where, as in Henry, the court is concerned with prior testimony, as opposed to out of court, unsworn, statements. Testimony is advanced for its truth, so if a witness had previously given evidence to one effect, it seems appropriate to use that on a later trial as evidence of its truth where it conflicts with what the witness now testifies.

In Australia, s 60 of the uniform evidence Acts has been interpreted to allow prior testimony to be used as proof of its truth, at least as far as intended, as opposed to unintended, assertions are concerned (Lee v R (1998) 195 CLR 594 (HCA)).

In New Zealand, clause 33 of the Evidence Bill sets out proposed "truthfulness rules", including a rule that the rules about hearsay evidence "do not apply to exclude evidence about reputation that relates to truthfulness". This formula, "do not apply to exclude", appears to mean that such evidence may be admissible for a hearsay purpose, that is, as evidence of its truth. The position here, however, is far from clear. If a court transcriber (judge's associate) were to be called as a witness to give evidence of what the accused had said at an earlier trial, to rebut what the accused has said at the present trial, that would not be hearsay evidence as defined in the Bill, because it would be evidence about a previous statement by a witness (see cl 4 definition of "hearsay statement"). Again, while this evidence would be admissible on the issue of credibility, it is not clear whether it is also admissible for the truth of what the accused previously said.

Currently, s 10 of the Evidence Act 1908[NZ] covers the position, insofar as it permits witnesses to be cross-examined about their previous inconsistent statements. Judicial opinion on whether the previous inconsistent statements admitted under this provision are evidence of their truth has varied from the mid-nineteenth century. Cross on Evidence, para 9.53, states that the present law is that they are not proof of their truth, citing Tepaki v Police [1967] NZLR 337, and at para 9.54 the learned authors say that the rationale for this is "simply another instance of the pseudo-logic occasionally indulged in by lawyers", while they acknowledge that previous statements made on oath may raise different issues than those made under differing circumstances.

Monday, December 12, 2005

The "rule" in Browne v Dunn

Can the rules of evidence, that have been designed to ensure fairness, be ignored if the judge feels that it would be more fair to ignore them? Certainly not, says the High Court of Australia in MWJ v R [2005] HCA 74 (7 December 2005).

In MWJ an inconsistency emerged with evidence that had been given by the complainant, when a subsequent witness gave evidence. Could the accused rely on the apparent inconsistency to undermine the complainant’s evidence? The inconsistency was "apparent" because the complainant was not re-called for questioning on the topic to see if the inconsistency was maintained.

This raised what is referred to in general terms as the rule in Browne v Dunn (1893) 6 R 67, which is that fairness usually requires that if a witness’s testimony is to be disputed, the grounds for that dispute should be put to that witness. The High Court noted that this rule has been applied in criminal law in many jurisdictions, including New Zealand, citing Gutierrez v R [1977] 1 NZLR 192, (1996) 14 CRNZ 108 (CA). In that case, the New Zealand Court of Appeal held

" … the rule is simply one of fairness. Has a reasonable opportunity been given to enable the evidence in question to be properly assessed? It is the responsibility of prosecuting counsel or a prosecutor who proposes to attack the credibility of defence witnesses, including the defendant, to cross-examine in a way which makes it plain that the relevant evidence is challenged and gives the witness a fair opportunity to answer the challenge. Such cross-examination however may not be necessary if from what has gone before or from the circumstances of the case it is fairly made plain that the truthfulness of particular facts given in evidence is not accepted, and an adequate opportunity to meet the challenge has otherwise been afforded.

"If evidence relevant to credibility is not so tested when it ought to be, it is likely to be unreasonable for the trier of fact to make an adverse finding in respect of credibility."

The High Court of Australia was unanimous in MWJ that in the circumstances of the case there was no miscarriage of justice and the appeal against conviction was dismissed. Two judgments were delivered, and they differ slightly on the point of when it is appropriate to give weight to the apparent inconsistency in the evidence.

Gleeson CJ and Heydon J held (para 19):

"In the present case, there was no obligation on trial counsel for the appellant to question the complainant about whether there had been more than one incident of sexual abuse at Sutcliffe Street, and there was no obligation to seek to have the complainant recalled for that purpose. Why would counsel for the appellant want to run the risk of eliciting further evidence of uncharged criminal acts by her client? That, no doubt, left the trial judge in a difficult position when he came to evaluate a criticism (in final address) of the complainant's credibility based on the supposed (although, in truth, non-existent) inconsistencies. It did not mean that counsel could not put her argument to the judge. As Doyle CJ said, it was a matter to be taken into account in assessing the weight to be given to the supposed inconsistencies. In the event, it was the fact that counsel chose (with reason) to leave the evidence in a state of uncertainty that undermined her submission about inconsistency. That was a forensic choice for counsel to make."

The penultimate sentence of this extract has difficulties. It was not for counsel for the accused to ask for re-call of the complainant; it was for counsel for the prosecution to offer (or, more accurately, to apply for) re-call of the witness. If counsel for the defence opposed that, less weight might have attached to subsequent reliance on inconsistencies. This is how the other Judges in MWJ, Gummow, Kirby and Callinan JJ, saw it (para 41):

"…As soon as the inconsistency emerged, and the trial judge rejected the appellant's objection to the evidence intended to be adduced from the complainant's mother [whose evidence had appeared to be inconsistent with that of her daughter, the complainant], it was open for the prosecution to offer to tender the complainant for further cross-examination. Had that happened it would then, and only then have been for the appellant, to decide whether to embrace the offer or not. If he had not, then and only then would the criticism that the Court of Criminal Appeal made of his conduct have been valid."

This seems to be the better view, but one might still wonder whether it is really accurate. If it would be unfair to permit recall of the complainant in view of, for example, the obligation on the prosecution not to split its case, then it would be unfair to ignore the apparent inconsistency. Each case will turn on its own facts, but fairness is the all-pervasive consideration.

What, then, if the rules of evidence appear to operate unfairly? This point was addressed in the three-judge joint judgment, para 37:

"Something should first be said of the trial judge's criticism of the appellant's failure, in effect, to give the complainant an opportunity of explaining away the inconsistency arising out of her mother's evidence. The criticism is ill-founded for these reasons. The complainant had already given her evidence when the mother gave her evidence. It was not for the appellant to know and anticipate, by cross-examining the complainant, what the mother would say about the complainant's assertions of complaints of multiple offences at Sutcliffe Street. It was not for the appellant to iron out inconsistencies in the case for the prosecution. Secondly, his Honour erred in holding that if there were competition between the avoidance of unfairness to the complainant and a "technical view of the rules of evidence" … (whatever that in the circumstances means), the former must prevail. It is not for a judge to depart from the rules of evidence on such a basis. The rules are designed to ensure fairness to all, certainly not least, to an accused in a criminal trial."

The difficulty in this apparent contrast between rules and fairness, is that the so-called rule in question is itself more of a principle than a rule. It is, as the New Zealand Court of Appeal said in Gutierrez, above, one of fairness. While it is usually called the "rule" in Browne v Dunn, that is simply shorthand non-technical language.

Friday, December 09, 2005

How civilised are we?

What use of torture is tolerable? More precisely, what risk that evidence tendered to the court was obtained by torture is tolerable? The House of Lords has answered, if on the balance of probabilities there was no torture, that is acceptable: A v Secretary of State for the Home Department [2005] UKHL 71 (8 December 2005).

This means that a risk of torture, just falling short of tipping the balance, can be ignored. It is notable that this case involved 7 Law Lords, and they split 4-3 on this difficult issue. Of course, they unanimously held torture to be repugnant to the common law. The minority, Lords Bingham, Nicholls and Hoffmann, held that evidence was inadmissible if the court was left with a real risk that it had been obtained by torture.

We should remember that in R v McCuin [1982] 1 NZLR 13 (CA) it was held that statements by accused persons should be excluded unless the court was sure beyond reasonable doubt that they had been made voluntarily.

As we often see in cases considered in this blog over the last 16 months, the choice between formalist and pragmatic reasoning is fundamental. This is illustrated in A’s case by Lord Hope, para 119:

"… it would be unrealistic to … demand that each piece of information be traced back to its ultimate source and the circumstances in which it was obtained investigated so that it could be proved piece by piece, that it was not obtained under torture. The threshold cannot be put that high. Too often we have seen how the lives of innocent victims and their families are torn apart by terrorist outrages. Our revulsion against torture, and the wish which we all share to be seen to abhor it, must not be allowed to create an insuperable barrier for those who are doing their honest best to protect us. A balance must be struck between what we would like to achieve and what can actually be achieved in the real world in which we all live. Articles 5(4) and 6(1) of the European Convention, to which Lord Bingham refers in para 62, must be balanced against the right to life that is enshrined in article 2 of the Convention."

Lord Carswell reasoned as follows (para 158):

"After initially favouring the Bingham test, I have been persuaded that the Hope test should be adopted … in determining whether statements should be admitted when it is claimed that they may have been obtained by the use of torture. Those who oppose the latter test have raised the spectre of the widespread admission of statements coming from countries where it is notorious that torture is regularly practised. This possibility must of course give concern to any civilised person. It may well be, however, that the two tests will produce a different result in only a relatively small number of cases …. Moreover, as my noble and learned friend Lord Brown of Eaton-under-Heywood points out in para 166 of his opinion, intelligence is commonly made up of pieces of material from a large number of sources, with the consequence that the rejection of one or some pieces will not necessarily be conclusive. While I fully appreciate the force of the considerations advanced by Lord Bingham in paras 58 and 59 of his opinion, I feel compelled to agree with Lord Hope's view in para 118 that the test which he proposes would, as well as involving fewer practical problems, strike a better balance in the way he there sets out."

Reflecting on this case, one is tempted to think that, as the courts have to respond to the difficult issues raised by the State’s responses to terrorism, a law peculiar to the demands of this sort of warfare will develop. It may well be that where public safety is a major consideration, a criterion for admissibility set at the level of the balance of probabilities can be acceptable, whereas in criminal cases of the ordinary kind the McCuin approach will remain appropriate.

Another point, on which laws about terrorism may properly set themselves apart, is the extent to which investigation of crime may be lawful, even though it is based on information obtained through torture. In ordinary criminal cases, the lawfulness of investigatory techniques, such as the use of interception devices, is assessed by ignoring information that had been improperly obtained. For example, if there is insufficient information available to support the granting of an interception warrant once improperly obtained information is ignored, the warrant is invalid: R v Crowe [1996] 3 NZLR 415; (1996) 13 CRNZ 708 (CA). But terrorism raises different values. As Lord Bingham said:

"68. … If the police were to learn of the whereabouts of a ticking bomb it would be ludicrous for them to disregard this information if it had been procured by torture. No one suggests the police should act in this way. Similarly, if tainted information points a finger of suspicion at a particular individual: depending on the circumstances, this information is a matter the police may properly take into account when considering, for example, whether to make an arrest.
"69. In both these instances the executive arm of the state is open to the charge that it is condoning the use of torture. So, in a sense, it is. The government is using information obtained by torture. But in cases such as these the government cannot be expected to close its eyes to this information at the price of endangering the lives of its own citizens. Moral repugnance to torture does not require this."


Judicial supervision of executive acts, however, can produce what in A was called a "mismatch" between information properly considered by the executive, and information properly admissible in evidence. While the Law Lords considered this mismatch was an acceptable inevitability (Lord Bingham at para 48 said "The common law is not intolerant of anomaly"), one might wonder how persistent this mismatch will be; this, indeed, is the central question. The law tends to adopt pragmatic compromises, as Lord Bingham pointed out in para 16:

" … there is an obvious anomaly in treating an involuntary statement as inadmissible while treating as admissible evidence which would never have come to light but for the involuntary statement. But this is an anomaly which the English common law has accepted, no doubt regarding it as a pragmatic compromise between the rejection of the involuntary statement and the practical desirability of relying on probative evidence which can be adduced without the need to rely on the involuntary statement."

On another point, but one that arises from this sort of legislative endorsement of illegal investigatory procedures, Lord Bingham let slip an old fashioned remark. At para 51 he said:

"… it would of course be within the power of a sovereign Parliament (in breach of international law) to confer power … to receive third party torture evidence."

Well, would it? Historical understanding of sovereignty supports this view, but the development of strong international law, with which this case is largely concerned, is incompatible with the notion that Parliament can make any law whatsoever. Further, the implications of Lord Bingham’s remark are that the British people may embrace barbaric behaviour, condemning themselves in the same terms as the Law Lords, and all civilized people, condemn torture.

Similar issues, to those considered in A's case, could arise in New Zealand under the Terrorism Suppression Act 2002, which (to summarise some of its aspects) allows the Prime Minister to use any relevant information to designate as a terrorist any person or organisation (s 30), and which creates offences of involvement with terrorists (s 13). On prosecution, a dispute may arise over the applicability of the definition to a particular person or organisation, and some inquiry into the source of the Prime Minister's information may be necessary: s 39(4) "In determining the proceedings, the Court may take into account any relevant classified security information available to it, even though that information has not been disclosed to or responded to by other parties to the proceedings."

Monday, December 05, 2005

Synchronicity

The House of Lords and the New Zealand Supreme Court decided similar cases on 1 December 2005. Each concerned the rights of persons who had been convicted, and each involved sentencing. They differed, in general terms, in that the English case concerned a transitional question arising from a legislation change, whereas the New Zealand case concerned the effect of a change in the circumstances of the offender, namely his age, on his eligibility for the indeterminate sentence of preventive detention. Both cases were decided by applying considerations of fairness.

In R (on application of Hammond) v Secretary of State for the Home Department [2005] UKHL 69 (1 December 2005) a literal reading of the relevant legislation required that the punitive period of a sentence of life imprisonment must be set by a High Court Judge on the papers and without an oral hearing. It was held that this should be read as being subject to an implied condition that an oral hearing must be permitted where that is necessary to comply with the prisoner’s right to a fair trial under art 6(1) of the European Convention on Human Rights.

Interestingly, this implication of an exception to an absolute legislative provision was accepted by the Secretary, once it had been decided that the absolute provision breached the Convention. There was, accordingly, no need to reconsider this approach to interpretation, which had been established in R v A [2002] 1 AC 45 (HL), where it was held that such implication was required by s 3(1) of the Human Rights Act 1998[UK], which provides:

"So far as it is possible to do so, primary legislation and subordinate legislation must be read and given effect in a way which is compatible with the Convention rights."

International law was also relevant in Mist v R [2005] NZSC 77 (1 December 2005). Here, preventive detention could be imposed on an offender, who was otherwise eligible, if he was "not less than 21 years of age". The provision did not specify whether this age was to be taken as at the date of the commission of the offence, or at the date of sentencing.

The Court was unanimous that the answer was the age of the offender at the time of the offence. Three Judges decided this on a narrow point: the interpretation of s 4(2) of the Criminal Justice Act 1985 (legislation which has since been replaced with provisions clear on the point). The other two Judges, Elias CJ and Keith J, agreed with that, and added broader considerations, which included the need to read s 4 consistently with art 15 of the International Covenant on Civil and Political Rights. Gault J summarised the opposing arguments in a way that suggested they were finely poised, but at para 62, he expressed agreement with a generous interpretation of s 4(2) "in conformity", apparently meaning the requirement of consistency with rights to which Elias CJ and Keith J had referred.

Eilas CJ and Keith J referred to the need for fairness and due process (paras 28, 29) and to developments in human rights law over the last 30 years (para 39), and to the need to give individuals the full measure of their rights (para 45).

Implicit in this approach is compliance with s 6 of the New Zealand Bill of Rights Act 1990, which is as follows:

"Wherever an enactment can be given a meaning that is consistent with the rights and freedoms contained in this Bill of Rights, that meaning shall be preferred to any other meaning."

The Bill of Rights is, according to its statement of purpose, "An Act – … (b) To affirm New Zealand’s commitment to the International Covenant on Civil and Political Rights".

Article 15(1) of that Covenant states:

"No one shall be held guilty of any criminal offence on account of any act or omission which did not constitute a criminal offence, under national or international law, at the time when it was committed. Nor shall a heavier penalty be imposed than the one that was applicable at the time when the criminal offence was committed. If, subsequent to the commission of the offence, provision is made by law for the imposition of the lighter penalty, the offender shall benefit thereby."

The first sentence of this is echoed in s 26(1) NZBORA. The second and final sentences are reflected in s 25(g) of NZBORA, although the wording differs slightly:

"(g) The right, if convicted of an offence in respect of which the penalty has been varied between the commission of the offence and sentencing, to the benefit of the lesser penalty."

Of course, in Mist it was the offender’s age, not the legislation, that had changed, but Elias CJ and Keith J, and implicitly Gault J, were prepared to apply the applicable provision consistently with art 15.

Significantly, this approach does not involve a close reading of s 6 NZBORA (which, indeed, was not cited in Mist). There was no agonizing over how strongly s 6 compelled an interpretation consistent with rights, and no attempt to contrast the approach required by s 6 NZBORA with the approach required of English courts by s 3 of the Human Rights Act 1998[UK]. The force of international rights law was recognised by three judges as relevant to the interpretation problem in Mist. As in Hammond, the rights-mandated approach to interpretation was accepted, without the need for argument over the strength of the legislative direction to conform to rights.

This has relevance to the forthcoming issue of the standard of proof imposed on the defence by reverse-onus provisions, which is essentially the question on which leave to appeal was granted in R v Hansen [2005] NZSC 74 (24 November 2005). Developments in international rights law have revealed an aversion to interpreting reverse onus provisions as placing legal burdens on the defence. Those developments can be followed in New Zealand without straining the words of s 6 of NZBORA.

Thursday, November 24, 2005

Bain, again

Supporters of David Bain will, no doubt, be studying Mallard v R [2005] HCA 68 (15 November 2005). Both started as petitions for the exercise of the Royal prerogative. Just as in R v Bain [2004] 1 NZLR 638, (2003) 20 CRNZ 637 (CA), this was a reference to the court of the question of the conviction of the petitioner, although, in Mallard, the proceedings reached the High Court of Australia on appeal from the determination of the reference by the Court of Criminal Appeal of Western Australia. The equivalent New Zealand provision is s 406 of the Crimes Act 1961. In Bain, the Court of Appeal noted, para 4, that

"A reference under s 406(a) has the effect of an appeal against the convictions so referred. Hence this Court must consider the matters arising as if David Bain was appealing against his convictions a second time."

In Mallard, the corresponding provision was described, at para 6, in the following terms:

"The significance of this history for present purposes, is that the exercise for which s 140(1)(a) of the Sentencing Act 1995 [WA] provides is effectively both a substitute for, and an alternative to, the invocation, and the exercise of the Crown prerogative, an exercise in practice necessarily undertaken by officials and members of the Executive, unconfined by any rules or laws of evidence, procedure, and appellate conventions and restrictions. That history, briefly stated, argues in favour of an approach by a court on a reference of a petition by the Attorney-General to it, of a full review of all the admissible relevant evidence available in the case, whether new, fresh or already considered in earlier proceedings, however described, except to the extent if any, that the relevant Part of the Act may otherwise require."

A significant point concerning the approach of the appellate court in these circumstances (where fresh evidence is relied on by the petitioner) was made in Mallard. This is that it is wrong for the appellate court to view the evidence in a way that is constrained by the verdict that had been reached at trial. At para 10 of Mallard the Court criticised the approach taken by the CCA (WA) whereby limitations were perceived on the court’s jurisdiction to consider the evidence. Instead, the HCA held that the reference of "the whole case" to the appellate court carried no such limitations, and

"…The inhibitory purpose and effect of the words "as if it were an appeal" are merely to confine the Court to the making of orders, and the following of procedures apposite to an appeal, and further, and perhaps most relevantly, to require the Court to consider whether the overall strength of the prosecution case requires the Court to apply the proviso …".

The corresponding New Zealand provision does not use the phrase "the whole case", but it is clear that the meaning is intended to be the same. It refers to "the question of the conviction", and, as with the Western Australian provision, it contrasts that with reference of "any point arising in the case".

It seems fair to say, therefore, that on reference of the conviction, the Court of Appeal should not feel constrained in its view of the facts by the jury’s decision. It is wrong, on this view, to say that because the jury must have accepted certain evidence, that that evidence carries enhanced status for the purposes of determining the reference.

Another indication of the correct approach, stated in Mallard at para 23, is

"It was not for the Court of Criminal Appeal to seek out possibilities, obvious or otherwise, to explain away troublesome inconsistencies which an accused has been denied an opportunity to explore and exploit forensically."

In a separate, concurring, judgment, Kirby J held, para 84, that in a fresh evidence case, the question was whether absence of that evidence

"… could have seriously undermined the effective presentation of the defence case, [as] a verdict reached in the absence of the material evidence (and the use that the defence might have made of it) cannot stand."

Mallard also contains a useful review of the prosecution’s duty of disclosure of evidence to the defence.

In the light of Mallard, the need to reconsider Bain is clear.

Wednesday, November 16, 2005

Death and certainty

How much formalism is too much? In interpreting statutes, to resolve an ambiguity, judges may focus on the context of the problematic phrase: the paragraph, subsection, and the placing of the section in the scheme of the Act, are considered. In addition, standard tools of interpretation, such as ejusdem generis, and noscitur a sociis, may be used.

Another approach, which may be used in conjunction with this formalistic method, or as an alternative, is the purposive, or pragmatic approach. According to this, the meaning of ambiguous legislation should be determined by reference to what would promote or further Parliament’s purpose in enacting the provision.

There is no consistency in approach to interpreting ambiguous criminal legislation. For example, in R v Secretary of State for the Home Dept; ex p Pierson [1998] AC 539, Lord Steyn said at 585:

"Counsel for the Home Secretary argued that the fixing of the tariff cannot be a sentencing exercise because the judge pronounces the only sentence, ie one of life imprisonment. This is far too formalistic. In public law the emphasis should be on substance rather than form. This case should also not be decided on a semantic quibble about whether the Home Secretary’s function is strictly ‘a sentencing exercise’. The undeniable fact is that in fixing a tariff in an individual case the Home Secretary is making a decision about the punishment of the convicted man."

And, in R v Karpavicius [2002] UKPC 59 (PC) Lord Steyn said, para 15:

"…In a more literalist age it may have been said that the words of s 6(2A)(c) [of the Misuse of Drugs Act 1975[NZ]] are capable of bearing either a wide and narrow meaning and that the fact that a criminal statute is involved requires the narrower interpretation to be adopted. Nowadays an approach concentrating on the purpose of the statutory provision is generally to be preferred: Cross, Statutory Interpretation (3rd ed), 1995, pp 172-175; Ashworth, Principles of Criminal Law (3rd ed), 1999, pp 80-81. This is reinforced by s 5(1) of the Interpretation Act 1999 [New Zealand] which provides that the meaning of an enactment must be ascertained from its text and in the light of its purpose."

We must, then, wonder whether Lord Steyn has consistently taken this pragmatic approach in Smith v The Queen (Jamaica) [2005] UKPC 43 (14 November 2005). Here, the Privy Council split 3-2 on a question of interpretation, and Lord Steyn was part of the majority.

The issue in Smith was the meaning of the phrase "in the course or furtherance of" in the definition of capital murder in Jamaica. There, capital murder (ie where the sentence may be death) occurs where murder is committed "in the course or furtherance of … burglary or housebreaking" (s 2(1)(d) of the Offences Against the Person Act 1864, as amended in 1992).

The facts of Smith were refreshingly brief; they were summarised in the minority judgment, para 18, as "The appellant stood on a ladder against the outside of the deceased's house at night, pulled aside a curtain and a piece of plastic in a window, inserted his head and upper body through the window and struck the deceased a number of blows with a machete as she lay in her bed under the window."

There was no doubt that his entry with felonious intent constituted burglary. The question was whether the murder was in furtherance of the burglary. He committed the burglary with intent of murdering. The minority considered that the murder was therefore committed in the course of or in furtherance of the burglary. This, one might have thought, is the ordinary and natural meaning of the phrase. But no; another interpretation is possible: the offender had a single purpose of murder, and the burglary was in furtherance of the murder, not the other way around, so the murder was not within the definition of capital murder. This was the majority view.

There are two competing public policy justifications. The minority refer, para 23, to the purpose of the section as being:

"…to protect citizens from being murdered in their own homes by intruders who break in at night and to deter offenders from committing such murders. We consider that the legislature could not have intended that an intruder who broke into a house, which he believed to be unoccupied, for the purpose of stealing therein and then, coming upon the occupier, killed him or her, should be guilty of capital murder, but that a person who broke into a house with the express purpose of killing the occupant and did so should not be guilty of capital murder."

On the other hand, the majority refer, para 8 to:

"The vice in these cases, which was thought by …[Parliament] to justify the death penalty, was that the defendant resorted to killing his victim in the course or furtherance of committing the [lesser offence]. It was the wanton and cynical nature of the killing, the debasing in the context of a comparatively minor criminal act of the value that is to be attached to human life, that was regarded as particularly reprehensible."

The majority adhered to authority that required a "duality of purpose" before capital murder was committed: Lamey v The Queen [1996] 1 WLR 902 (PC). Such adherence to precedent is an illustration of formalism being dominant over pragmatism: see Thomas, "The Judicial Process" (2005). Thomas would remove all formalism from the law, whereas the current approach of courts is to employ either formalism, pragmatism, or a blend of them both, as seems appropriate.


This is also illustrated in Lamey, where the Board (which again included Lord Steyn) considered the interpretation of another paragraph of s 2(1) of the Jamaican legislation. This one, (f), includes within the definition of capital murder, any murder committed in furtherance of an act of terrorism. The unanimous judgment refers to the approach to penal statutes:

"6. The starting point in any consideration of section 2(1)(f) must be the fact that its object was to reduce the categories of murder which attracted the death penalty. It follows that a construction which produces little or no reductive effect is unlikely to be correct. Furthermore regard must be had to the general principle that a person should not be penalised and in particular should not be deprived of life or freedom unless under clear authority of law (Bennion's Statutory Interpretation, 2nd Edn. page 574)."

Note the slightly more restrictive approach endorsed here, compared with that taken in Karpavicius, above. In Lamey the conclusion was that paragraph (f) required two intents:

"8. … In their Lordships' view the paragraph requires there to be a double intent on the part of the murderer namely an intent to murder and an intent to create a state of fear in the public or a section thereof."

In Smith, this approach was applied to paragraph (d), and this reasoning is an example of formalist analysis.

Monday, November 07, 2005

Dealing with vagueness

Rights may be expressed in unavoidably vague terms, so how can they be said to be absolute? An example is the right not to be subjected to inhuman or degrading treatment. Does it make sense to say that this right is absolute?

This was the fundamental question that confronted the House of Lords last week in R v Secretary of State for the Home Department, ex parte Adam [2005] UKHL 66 (3 November 2005). It was decided that this right is absolute, once a threshold level of inhuman or degrading treatment has been reached, and that the threshold is relative to the circumstances of each case. Furthermore, the threshold is a high one (except where pain and suffering are deliberately inflicted).

It may be clearer to say that, once treatment can properly be described as inhuman or degrading, then there has been a breach of the right. To say that the threshold is a high one (in particular, Lord Bingham, para 7) is really a way of saying that inhuman or degrading treatment must be relatively harsh. How harsh, depends on the circumstances. On this point, Lord Hope quoted, at para 54, cases from the European Court of Human Rights:

"But the European Court has all along recognised that ill-treatment must attain a minimum level of severity if it is to fall within the scope of the expression "inhuman or degrading treatment or punishment": Ireland v United Kingdom (1978) 2 EHRR 25, 80, para 167; A v United Kingdom (1998) 27 EHRR 611, 629, para 20; V v United Kingdom (1999) 30 EHRR 121, para 71. In Pretty v United Kingdom 35 EHRR 1, 33, para 52, the court said:

‘As regards the types of 'treatment' which fall within the scope of article 3 of the Convention, the court's case law refers to 'ill-treatment' that attains a minimum level of severity and involves actual bodily injury or intense physical or mental suffering. Where treatment humiliates or debases an individual showing a lack of respect for, or diminishing, his or her human dignity or arouses feelings of fear, anguish or inferiority capable of breaking an individual's moral and physical resistance, it may be characterised as degrading and also fall within the prohibition of article 3. The suffering which flows from naturally occurring illness, physical or mental, may be covered by article 3, where it is, or risks being, exacerbated by treatment, whether flowing from conditions of detention, expulsion or other measures, for which the authorities can be held responsible.’

It has also said that the assessment of this minimum is relative, as it depends on all the circumstances of the case such as the nature and context of the treatment or punishment that is in issue. The fact is that it is impossible by a simple definition to embrace all human conditions that will engage article 3."


Significantly, the House of Lords in Adam rejected the spectrum analysis applied in the Court of Appeal, pursuant to which the question whether inhuman or degrading treatment amounts to a breach of the right depends on whether it is nevertheless justified, for example by government policy. Lord Brown, while not actually approving the spectrum analysis, came close to applying it by reasoning that the motive for the treatment may be relevant (para 94). Lord Hope, while not expressly distancing himself from Lord Brown’s approach, pointed out, at para 55, that it would be wrong to include government policy in the assessment of the threshold as that would allow proportionality in by the back door.

Given, then, that the concept of inhuman or degrading treatment is not qualified by considerations of government policy, how close can we get to a definition of it? Lord Bingham put it like this, para 8-9:

"… The answer must in my opinion be: when it appears on a fair and objective assessment of all relevant facts and circumstances that an individual applicant faces an imminent prospect of serious suffering caused or materially aggravated by denial of shelter, food or the most basic necessities of life. Many factors may affect that judgment, including age, gender, mental and physical health and condition, any facilities or sources of support available to the applicant, the weather and time of year and the period for which the applicant has already suffered or is likely to continue to suffer privation.
[9] It is not in my opinion possible to formulate any simple test applicable in all cases. But if there were persuasive evidence that a late applicant was obliged to sleep in the street, save perhaps for a short and foreseeably finite period, or was seriously hungry, or unable to satisfy the most basic requirements of hygiene, the threshold would, in the ordinary way, be crossed."


We might like to consider whether this approach is applicable, by analogy, to the determination of fair trial issues. Whether the appropriate circumstances amount to unfairness can depend on the assessment of factors such as whether the accused would be able properly to challenge the prosecution case, and whether the case would be determined by an unbiased tribunal. Once that arises (not: once that threshold is crossed), then there is a breach of the absolute right to a fair trial.

Leaving readers with that thought, I return to celebrating the 25th anniversary of the publication of my first article, which appeared in England, in what was then the leading criminal journal in the common law world, the Criminal Law Review.

Tuesday, October 25, 2005

Whose verdict?

Appeals against conviction require consideration of whether there has been a substantial miscarriage of justice. This, in turn, can raise two questions: whether the trial was fair, and, whether the accused was wrongly deprived of a real chance of an acquittal. These are independent questions. The latter is often (and wrongly) taken to be an opportunity for an exacting judicial analysis of the evidence, notwithstanding the oft-repeated claim that it is not for the appellate judges to substitute their verdict for that of the jury. R v Bain [2004] 1 NZLR 638, (2003) 20 CRNZ 637 (CA) is a prime example of applying the trained judicial mind to the analysis of the evidence, rather than considering how the jury might have reacted to proposed new evidence.

An interesting observation on the fact that jurors may not apply linear logic to their task of coming to a verdict was recently made by Kirby J (happily siding with the majority) in R v Stevens [2005] HCA 65 (21 October 2005), para 82:

"One assumes that the human mind, and even more the collective mind of a jury, operates in serious decision-making, rationally and reasonably. But the mind does not necessarily act according to linear paths of strict logic. At any time in a criminal trial, several issues are in play. As Callinan J correctly points out [at para 158: "Nor do I think it is an answer in this case to say that one defence, or a direction in respect of it, subsumed another to the extent that the latter needed not to be mentioned or put to the jury in appropriate terms. Different people may have different perceptions of facts. Certain words, or language, or expressions of concepts, may provoke different responses in different people"], different people, especially a group of people, may have different perceptions of facts and of words, expressions and language (such as on being told of the substance of the Code's provisions on accident). The appellant, who was facing, if convicted, the heaviest penalty known to the law, was entitled to have the chance of a favourable response of the jury to the exemption provided by the Code from criminal responsibility for accident, properly explained. The trial judge ought not to have deprived the appellant of that chance."

An example of acute judicial analysis of the trial evidence, this time with a view to discovering whether the accused had been deprived of the right to adequate facilities to prepare a defence, is the judgment of Thomas J in R v Griffin [2001] 3 NZLR 577, (2001) 19 CRNZ 47 (CA). He concluded that on the evidence, lack of opportunity to have a defence expert examine the complainant did not adversely affect the result, and therefore there was no unfairness. That was, with respect, an incorrect melding of the two questions set out above. He may well have been correct to conclude that there had been no loss of a real chance of an acquittal, on the evidence adduced at trial. But, if the trial was unfair there should have been (as the majority held) a substantial miscarriage of justice. Thomas J treated procedural fairness as irrelevant if the verdict appeared to be correct. He allowed pragmatism (or, as he calls it, substantialism) to override the formal requirements of a fair trial. The majority of the Privy Council did the same in R v Howse [2005] UKPC 31 (19 July 2005), blogged here on 23.7.05. In Griffin, the majority held that had it could not be said that, had the error not occurred, the jury would inevitably have convicted, and a new trial was ordered. They did, however, recognise that this consideration of the second question was probably unnecessary, as they had found that the trial had been unfair. They said, para 40, that it was difficult to imagine a case where there would not be a substantial miscarriage of justice after a breach of the relevant right.