Should one feel sorry for a person who, having served a sentence of imprisonment, has his convictions quashed on appeal because the evidence against him should not have been ruled admissible?
A concurrent sentence of two years imprisonment for two aggravated robberies was served by the appellant "S", a 15 year-old. Then, on appeal, two High Court judges sitting together held that the evidence of his guilt, which was entirely contained in his statement to the police, should have been excluded, and his convictions were quashed: S v Police 14/3/06, Baragwanath and Heath JJ, HC Auckland CRI 2004-404-515.
In reality, unless his confessions were false, S was guilty and, no doubt, he deserved his sentence. But that is to ignore the rule of law, which in this context means that punishment can only be imposed following due process of law. If by "due process" we mean upon lawful conviction, then S did not receive due process.
The problem arises from practicalities. Sentences take effect before rights of appeal against convictions have been exercised. Trials proceed before rights of appeal against evidential rulings have been exercised. Evidential rulings during a trial must be left for appeal after completion of the trial, otherwise trials would have to adjourn to allow appeals to be heard.
The obvious course, to minimise "false" imprisonment, would be to defer sentencing hearings until appeals had been determined, but that is often impractical: many convicted people would run away, or would have to be denied bail.
If there is no viable alternative to the present procedure, should people like S be compensated? How would compensation be measured? Our assessment of the value of due process may have to be balanced against the harm caused by the offender. The courts may be forced to say that the real remedy is the vindication that follows from the quashing of the conviction: see further, my blogs of 6 and 30 March 2005.
In S v Police, a police officer had told S that he wanted him to point out where burglaries had occurred, so that files could be cleared up, and that he would not be charged with committing them. Unexpectedly, from the officer’s point of view, S said he had committed a couple of aggravated robberies. He was interviewed about those, he confessed to them, and was charged. If his initial mention of them had been a "spontaneous" utterance, his consequential dealings with the police would not have been flawed, as the correct procedures for dealing with young people had been followed thereafter. However, if the initial reference to the robberies was not spontaneous, in the sense that it was made in reliance on the officer’s assurance that he would not be charged, then it was made in circumstances that were in breach of the statutory procedures, and what followed, although correct in itself, could be tainted.
The High Court held that the utterance was not spontaneous, but rather it had been made in reliance on the officer’s assurance. There was an error of procedure surrounding that, so the question became whether the following procedures, which led to the making of the confession, were indeed tainted.
The Court found it necessary, at this point, to consider the nature of causation (para 56 – 67). This came down to asking whether, in the circumstances, the non-spontaneous utterance had been the "effective cause" of the subsequent confession. It had, and because the use of an inadmissible statement to secure a subsequent statement was contrary to the policy of the legislation (para 59), both were inadmissible.
A lot turned, in this case, on how the judges assessed the circumstances. It could easily have been decided that S had, on receiving the advice to which he was entitled, decided to make a clean breast of everything, and that that decision, rather than the earlier indication that he would not be charged, was the effective cause of his confession. In cases like this, where the assessment of the circumstances seems to be finely balanced, it is tempting (but, in law, wrong) to think that the court was influenced by the common sense solution: S had served his sentence and may as well be relieved of the convictions to give him some incentive to obey the law.
Did the absence of any compensation for the "wrongful" imprisonment facilitate this balancing of the circumstances?
Thursday, March 16, 2006
Tuesday, March 14, 2006
Inferences of innocence
A commonly advanced ground of appeal against conviction is that the evidence supported an inference of innocence and the judge did not correctly tell the jury how they should deal with inferences.
We have passed through a period when much attention was given to what inferences are and how they should be handled. Things got rather complex as efforts were made to be precise about this. The relationship between the process of drawing inferences, and the process of deciding whether something has been proved, was at the centre of this complexity.
Some facts are proved by direct evidence: for example, a witness says that he saw something happen. Other facts are inferred from direct evidence: the facts that are directly proved suggest that something else is true. In turn, these inferred facts may combine, with other inferred facts, or with directly proved facts, or with a mixture of inferred facts and direct facts, to enable a further inference that something else is true. Wherever an inference occurs, it is usually described as an ordinary process of logical thought.
As to proof, it is relatively easy to see that evidence of directly observed facts may prove those facts, to the necessary standard. If the fact is an element of the alleged offence, then, when the jury considers all the evidence in the case, it must decide whether that element has been proved beyond reasonable doubt. But, in relation to other facts, (and with some exceptions), no particular standard of proof is required. The jury does not have to be instructed by the judge about the standard to which they must be satisfied that inferred facts are proved before they can use them to support other inferences, such as an inference of an element of the offence, as long as, ultimately, the jury tests the proof of the elements of the offence against the standard of beyond reasonable doubt.
So, what about the inferences supporting innocence? In what has been regarded as the leading case on inferences in New Zealand, R v Puttick (1985) 1 CRNZ 644 (CA) the Court summarised the position:
"Inference is simply one of the mental processes which may be used by a jury in carrying out its primary task of assessing the evidence and deciding whether or not it establishes the guilt of the accused beyond reasonable doubt. Where the charge has several essential elements, proof of guilt necessarily involves proof of each of those elements to the same standard. It does not, however, require proof beyond reasonable doubt of every fact which may be relevant to proof of each essential element.
… It must be equally unhelpful to tell jurors that, if proven facts support two inferences of equal weight, they should accept one and reject the other. To draw an inference either way from such facts would be pure speculation. Jurors should not be directed to accept or reject inferences when they have no logical basis for either step."
This could suggest (although we can be sure the Court did not intend this reading) that an inference supporting innocence can neither be accepted nor rejected if it is equal in weight to an inference of guilt.
A clearer account has been given in R v Seekamut 10/7/03, CA82/03:
"If on an objective basis that has regard to all the circumstances, a rational alternative to guilt is not excluded, there must for that reason be a reasonable doubt. But the mere fact that some of the circumstances might arguably permit an inference inconsistent with guilt is not enough. The jury’s function is to assess the whole of the evidence and in so doing may conclude that a suggested alternative is not reasonably tenable."
Similarly, yesterday the Privy Council in Taylor v R (Jamaica) [2006] UKPC 12 (13 March 2006), para 18, held:
"Their Lordships agree with the submission made on behalf of the appellant that in the circumstances of this case it was essential that the judge … spell out the possible inferences to be drawn … and instruct them that they must rule out all inferences consistent with innocence before they could be satisfied that the inference of guilt has been proved correct."
This could be read as suggesting a slightly different approach to that in Seekamut: instead of treating the evidence as all being in a big pool, to be considered in the round to see what ends up being proved, the process alluded to in Taylor seems to be one of deciding first whether inferences of innocence can be ruled out, then, if they are, turning to see whether guilt has been proved. We might, however, reasonably wonder whether this explication is correct, insofar as it seems to place a burden of proof on the defence, and it also seems to make an artificial distinction between stages of the reasoning process. The correct approach would be to tell the jury that, if after considering the evidence, they are left with a reasonable doubt about the accused's guilt, they must find him not guilty.
Because the judge in Taylor failed properly to direct the jury in this regard, the appeal against conviction was allowed and the case remitted to determine whether there should be a retrial.
We have passed through a period when much attention was given to what inferences are and how they should be handled. Things got rather complex as efforts were made to be precise about this. The relationship between the process of drawing inferences, and the process of deciding whether something has been proved, was at the centre of this complexity.
Some facts are proved by direct evidence: for example, a witness says that he saw something happen. Other facts are inferred from direct evidence: the facts that are directly proved suggest that something else is true. In turn, these inferred facts may combine, with other inferred facts, or with directly proved facts, or with a mixture of inferred facts and direct facts, to enable a further inference that something else is true. Wherever an inference occurs, it is usually described as an ordinary process of logical thought.
As to proof, it is relatively easy to see that evidence of directly observed facts may prove those facts, to the necessary standard. If the fact is an element of the alleged offence, then, when the jury considers all the evidence in the case, it must decide whether that element has been proved beyond reasonable doubt. But, in relation to other facts, (and with some exceptions), no particular standard of proof is required. The jury does not have to be instructed by the judge about the standard to which they must be satisfied that inferred facts are proved before they can use them to support other inferences, such as an inference of an element of the offence, as long as, ultimately, the jury tests the proof of the elements of the offence against the standard of beyond reasonable doubt.
So, what about the inferences supporting innocence? In what has been regarded as the leading case on inferences in New Zealand, R v Puttick (1985) 1 CRNZ 644 (CA) the Court summarised the position:
"Inference is simply one of the mental processes which may be used by a jury in carrying out its primary task of assessing the evidence and deciding whether or not it establishes the guilt of the accused beyond reasonable doubt. Where the charge has several essential elements, proof of guilt necessarily involves proof of each of those elements to the same standard. It does not, however, require proof beyond reasonable doubt of every fact which may be relevant to proof of each essential element.
… It must be equally unhelpful to tell jurors that, if proven facts support two inferences of equal weight, they should accept one and reject the other. To draw an inference either way from such facts would be pure speculation. Jurors should not be directed to accept or reject inferences when they have no logical basis for either step."
This could suggest (although we can be sure the Court did not intend this reading) that an inference supporting innocence can neither be accepted nor rejected if it is equal in weight to an inference of guilt.
A clearer account has been given in R v Seekamut 10/7/03, CA82/03:
"If on an objective basis that has regard to all the circumstances, a rational alternative to guilt is not excluded, there must for that reason be a reasonable doubt. But the mere fact that some of the circumstances might arguably permit an inference inconsistent with guilt is not enough. The jury’s function is to assess the whole of the evidence and in so doing may conclude that a suggested alternative is not reasonably tenable."
Similarly, yesterday the Privy Council in Taylor v R (Jamaica) [2006] UKPC 12 (13 March 2006), para 18, held:
"Their Lordships agree with the submission made on behalf of the appellant that in the circumstances of this case it was essential that the judge … spell out the possible inferences to be drawn … and instruct them that they must rule out all inferences consistent with innocence before they could be satisfied that the inference of guilt has been proved correct."
This could be read as suggesting a slightly different approach to that in Seekamut: instead of treating the evidence as all being in a big pool, to be considered in the round to see what ends up being proved, the process alluded to in Taylor seems to be one of deciding first whether inferences of innocence can be ruled out, then, if they are, turning to see whether guilt has been proved. We might, however, reasonably wonder whether this explication is correct, insofar as it seems to place a burden of proof on the defence, and it also seems to make an artificial distinction between stages of the reasoning process. The correct approach would be to tell the jury that, if after considering the evidence, they are left with a reasonable doubt about the accused's guilt, they must find him not guilty.
Because the judge in Taylor failed properly to direct the jury in this regard, the appeal against conviction was allowed and the case remitted to determine whether there should be a retrial.
Monday, March 13, 2006
Metaphysical laws
Can the law have been other than it was understood to have been at the time?
If it can, are we currently operating under a misunderstanding of the law?
Are there points of law that we are mistakenly taking for granted?
If we all think that the law says one thing, can we be wrong?
Can some future court say we were wrong, and that the law said something else?
The first of these questions concerned the Privy Council last week in Bowe v R (Bahamas) [2006] UKPC 9 (8 March 2006). The issue involved detailed analysis of legislation, but in essence the question came down to whether the death penalty was mandatory in the Bahamas in 1973, notwithstanding that cases had been conducted on the basis that it was.
The Privy Council answered the conundrum this way (para 42):
"… it took some time for the legal effect of entrenched human rights guarantees to be appreciated, not because the meaning of the rights changed but because the jurisprudence on human rights and constitutional adjudication was unfamiliar and, by some courts, resisted."
Further, it would be wrong to answer the question by now trying to assess what chance the argument would have had of being accepted in 1973.
Instead,
"The task is to ascertain what the law, correctly understood, was at the relevant time, unaffected by later legal developments, since that is plainly the law which should have been declared had the challenge been presented then." (ibid)
So,
"It matters little what lawyers and judges might have thought in their own minds: in the context of a codified Constitution, what matters is what the Constitution says and what it has been interpreted to mean. In 1973 there was no good authority contrary to the appellants' argument, and much to support it. In the final resort, the most important consideration is that those who are entitled to the protection of human rights guarantees should enjoy that protection. The appellants should not be denied such protection because, a quarter century before they were condemned to death, the law was not fully understood." (ibid)
At this point, it seems to be only a subsidiary consideration that there was no authority to the contrary as at 1973, as the governing matter ("the final resort") is that those who are entitled to human rights protection should be entitled to enjoy that protection. Question-begging, of course, as the issue was whether the people were (not "are") entitled to that protection.
Not to worry, for in Bowe the death sentences were quashed because the law as at 1973 was that the sentence was discretionary, not mandatory.
This interpretation was therefore carried through to the present pursuant to the legislative provisions. The cases of the two appellants were remitted for consideration of the appropriate penalty.
It seems, consequently, that we are governed by laws that we have not yet discovered, and may not be discovered until we are all dead.
If it can, are we currently operating under a misunderstanding of the law?
Are there points of law that we are mistakenly taking for granted?
If we all think that the law says one thing, can we be wrong?
Can some future court say we were wrong, and that the law said something else?
The first of these questions concerned the Privy Council last week in Bowe v R (Bahamas) [2006] UKPC 9 (8 March 2006). The issue involved detailed analysis of legislation, but in essence the question came down to whether the death penalty was mandatory in the Bahamas in 1973, notwithstanding that cases had been conducted on the basis that it was.
The Privy Council answered the conundrum this way (para 42):
"… it took some time for the legal effect of entrenched human rights guarantees to be appreciated, not because the meaning of the rights changed but because the jurisprudence on human rights and constitutional adjudication was unfamiliar and, by some courts, resisted."
Further, it would be wrong to answer the question by now trying to assess what chance the argument would have had of being accepted in 1973.
Instead,
"The task is to ascertain what the law, correctly understood, was at the relevant time, unaffected by later legal developments, since that is plainly the law which should have been declared had the challenge been presented then." (ibid)
So,
"It matters little what lawyers and judges might have thought in their own minds: in the context of a codified Constitution, what matters is what the Constitution says and what it has been interpreted to mean. In 1973 there was no good authority contrary to the appellants' argument, and much to support it. In the final resort, the most important consideration is that those who are entitled to the protection of human rights guarantees should enjoy that protection. The appellants should not be denied such protection because, a quarter century before they were condemned to death, the law was not fully understood." (ibid)
At this point, it seems to be only a subsidiary consideration that there was no authority to the contrary as at 1973, as the governing matter ("the final resort") is that those who are entitled to human rights protection should be entitled to enjoy that protection. Question-begging, of course, as the issue was whether the people were (not "are") entitled to that protection.
Not to worry, for in Bowe the death sentences were quashed because the law as at 1973 was that the sentence was discretionary, not mandatory.
This interpretation was therefore carried through to the present pursuant to the legislative provisions. The cases of the two appellants were remitted for consideration of the appropriate penalty.
It seems, consequently, that we are governed by laws that we have not yet discovered, and may not be discovered until we are all dead.
Thursday, March 09, 2006
What a "fair trial" means
Today, the High Court of Australia decided Nudd v R [2006] HCA 9 (9 March 2006), a case with helpful dicta on the meaning of "fair trial". I have recently noted cases concerning what is a substantial miscarriage of justice (see Index), and have described the two ways in which miscarriages of justice may be considered "substantial" for the purposes of the proviso: by depriving the accused of a real chance of an acquittal, or by causing the trial to be unfair.
Nudd recognises that, even though the evidence of guilt may be overwhelming, there may be rare cases where there has nevertheless been a miscarriage of justice because the trial was rendered unfair to the accused: see Gleeson CJ at para 6, Kirby J at 87 and 100, Callinan and Heydon JJ at 158. The other judges, Gummow and Hayne JJ jointly decided that no miscarriage of justice had been demonstrated to have occurred (para 25). Gleeson CJ agreed with that conclusion.
Kirby J treated the outcome aspect (the depriving of the accused of a real chance of acquittal) in the same way as the other judges, holding that on any approach, there was no prospect of acquittal as the prosecution case was overwhelming, but he also gave the most detailed consideration to the other aspect, that of the fairness of the trial. He found this the more difficult point to decide (para 106), because the defects at trial (counsel’s serious incompetence, especially in not knowing the law and in failing to take detailed instructions and consequently to advise the client) were so serious. He held that this case was "borderline" on the question of fairness (para 109), but that the decisive point was that the evidence against the accused was conclusive of guilt.
This might be thought to muddy the distinction between the outcome aspect and the fairness aspect of substantial miscarriage of justice. However, I suggest that Kirby J’s reasoning is essentially that, from the point of view of fairness, the errors were inconsequential, because the defence in the trial had (and could only have) been to put the prosecution to proof, and the judge in her summing up had properly directed the jury on the law.
I have previously suggested that a fair trial is one where the law is accurately applied to the facts determined without bias. Applying this to Nudd, we can say that the errors at trial (counsel’s incompetence) did not cause the law to be applied inaccurately, nor did it cause the jury to be biased against the accused.
Nudd recognises that, even though the evidence of guilt may be overwhelming, there may be rare cases where there has nevertheless been a miscarriage of justice because the trial was rendered unfair to the accused: see Gleeson CJ at para 6, Kirby J at 87 and 100, Callinan and Heydon JJ at 158. The other judges, Gummow and Hayne JJ jointly decided that no miscarriage of justice had been demonstrated to have occurred (para 25). Gleeson CJ agreed with that conclusion.
Kirby J treated the outcome aspect (the depriving of the accused of a real chance of acquittal) in the same way as the other judges, holding that on any approach, there was no prospect of acquittal as the prosecution case was overwhelming, but he also gave the most detailed consideration to the other aspect, that of the fairness of the trial. He found this the more difficult point to decide (para 106), because the defects at trial (counsel’s serious incompetence, especially in not knowing the law and in failing to take detailed instructions and consequently to advise the client) were so serious. He held that this case was "borderline" on the question of fairness (para 109), but that the decisive point was that the evidence against the accused was conclusive of guilt.
This might be thought to muddy the distinction between the outcome aspect and the fairness aspect of substantial miscarriage of justice. However, I suggest that Kirby J’s reasoning is essentially that, from the point of view of fairness, the errors were inconsequential, because the defence in the trial had (and could only have) been to put the prosecution to proof, and the judge in her summing up had properly directed the jury on the law.
I have previously suggested that a fair trial is one where the law is accurately applied to the facts determined without bias. Applying this to Nudd, we can say that the errors at trial (counsel’s incompetence) did not cause the law to be applied inaccurately, nor did it cause the jury to be biased against the accused.
Thursday, March 02, 2006
Bad technique!
Sometimes, evidence that the accused has been convicted for committing other offences is admissible to prove that he committed the offence with which he is now charged. Or again, allegations that he committed other offences may be used to prove that he committed this one. Or, yet again, allegations of offences for which he has been acquitted may be used as evidence that he is guilty of this one.
This sort of evidence, called similar fact evidence, should only be allowed to be used in exceptional circumstances, for the obvious reason that it will have a tendency to bias the tribunal of fact (usually the jury) against the accused.
It must be said that the current law on similar fact evidence in New Zealand is in need of clarification. This is largely due to the decision of the Court of Appeal in R v Holtz [2003] 1 NZLR 667; (2002) 20 CRNZ 14 (CA), where the criterion for admissibility was held to be whether the probative value of the similar fact evidence outweighs its illegitimately prejudicial effect:
"One of the problems with evidence of past conduct, whether labelled evidence of bad character, propensity evidence or similar fact evidence, is the tendency to look for principles of admissibility applicable to all such evidence in all circumstances. At a very general level there is the broad principle that to be admissible the evidence must be such that its probative value outweighs illegitimate prejudice to the accused in having adduced evidence of past conduct that might be given undue weight or used improperly in reasoning towards guilt of the crime charged. It is in the application of that general principle in the circumstances of particular cases that difficulties arise. Often this is because the approach adopted in one set of circumstances is taken as the rule to be applied in a quite different case. Because of the wide variation in circumstances in which the issue calls for consideration, the preferable approach in a particular case is that which determines and weighs probative value and potential prejudice in the circumstances. It is necessary to consider the nature of the evidence, what it is sought to prove, what other evidence there is and its relationship and whether the evidence relates to prior proved offending or to concurrent charges, all against the underlying dangers inherent in propensity or bad character evidence." (para 35)
The so-called weighing of probative value against prejudicial effect is problematic conceptually, and has been criticised in the leading Australian case on similar fact evidence, Pfennig v R (1995) 182 CLR 461 per McHugh J on the basis that these were "incommensurables", and the Supreme Court of Canada has agreed with his criticism, in R v Handy (2002) 164 CCC (3d) 481 (SCC); 213 DLR (4th) 385, at paras 148-149. And I have written an article on it: "Probative value, illegitimate prejudice and the accused's right to a fair trial" (2006) 29 Crim LJ 8.
In Phillips v R [2006] HCA 4 (1 March 2006) the High Court of Australia has confirmed that Pfennig continues to be the authority on similar fact evidence, binding on Australian courts. It issued a thinly-veiled rebuke to courts that sought to find reasons for departing from Pfennig, at para 60:
"It must be said at once that it is for this Court alone to determine whether one of its previous decisions is to be departed from or overruled …. Of course, in criminal cases it is often necessary for trial judges and Courts of Criminal Appeal to elaborate upon rulings of this Court; to gather together rules expressed in several cases; to apply rules to different facts; and sometimes to reconsider rules affected by later legislation. Within spaces left by the binding determinations of this Court, trial judges and intermediate courts retain their proper functions …. However, these do not extend to varying, qualifying or ignoring a rule established by a decision of this Court. Such a rule is binding on all courts and judges in the Australian Judicature."
One misinterpretation of Pfennig, alluded to briefly in Phillips at para 62, is that it is appropriate and sufficient to regard similar fact evidence as a form of circumstantial evidence. This is what had occurred in Holtz. See Adams on Criminal Law, Ch2.8.19(5), and my blog entry for 14 October 2004. A danger of that view is that it leaves the process of drawing inferences unelaborated by particular directions to the jury, as authorities on inferences tend to avoid such assistance. It also tempts the court to admit the evidence, as circumstantial evidence is just a category of admissible evidence.
Phillips and Pfennig hold that similar fact evidence must pass a high threshold before it can be admitted (para 79):
"Criminal trials in this country are ordinarily focused with high particularity upon specified offences. They are not, as such, a trial of the accused's character or propensity towards criminal conduct. That is why, in order to permit the admission of evidence relevant to several different offences, the common law requires a high threshold to be passed. The evidence must possess particular probative qualities; a strong degree of probative force; a really material bearing on the issues to be decided. That threshold was not met in this case. It was therefore necessary that the allegations, formulated in the charges brought against the appellant, be separately considered by different juries, uncontaminated by knowledge of other complaints. This is what Pfennig and other decisions of this Court require. To the extent that O'Keefe [a Queensland Court of Appeal decision critical of the High Court of Australia's judgments in Pfennig] or other authority suggests otherwise, it does not represent the law. No other outcome would be compatible with the fair trial of the appellant."
Phillips had been charged with 8 counts of sexual offending, and had unsuccessfully sought severance of those. Joinder was permitted on the basis that the evidence on each charge was admissible on the others, as evidence of similar facts. The High Court summarised the evidence and its conclusion in this way (para 56):
"The similarities relied on were not merely not "striking", they were entirely unremarkable. That a male teenager might seek sexual activity with girls about his own age with most of whom he was acquainted, and seek it consensually in the first instance, is not particularly probative. Nor is the appellant's desire for oral sex, his approaches to the complainants on social occasions and after some of them had ingested alcohol or other drugs, his engineering of opportunities for them to be alone with him, and the different degrees of violence he employed in some instances. His recklessness in persisting with this conduct near other people who might be attracted by vocal protests is also unremarkable and not uncommon."
Bad technique, indeed, as it got Phillips to court.
The trial judge, and the Queensland Court of Appeal, had reasoned that so many similar complaints could not have arisen by chance, so together they had credibility. The High Court was not so slap-dash. It identified the issue on which the challenged evidence was relevant as being consent, and it pointed out that the non-consent of one complainant could not possibly prove that another had not consented.
One thing this case illustrates is the way judges can lose their reasoning powers in cases of alleged sexual misconduct. There is, in other words, a grave danger of judicial bias (probably unconscious bias) in such cases. In those circumstances, clear rules, rather than discretions, are needed concerning the admissibility of evidence of other alleged misconduct. The Pfennig rule is that similar fact evidence is not admissible unless there is no reasonable view of it that is consistent with the accused’s innocence. It must have strong probative force, a really material bearing on the issues to be decided; its probative force must clearly transcend its merely prejudicial effect; it must have a sufficient nexus, some specific connection, with the issues; its probative force must be such as to make it just to admit it (Phillips, para 54). None of those criteria were met in Phillips (para 55).
It is sad to note that in New Zealand, the Evidence Bill proposes an approach to similar fact evidence (or, in its terminology, "propensity evidence") that applies the probative value/prejudicial effect weighing, criticism of which is noted above. Clause 39(1) of the Bill provides:
"The prosecution may offer propensity evidence about a defendant in a criminal proceeding only if the evidence has a probative value in relation to an issue in dispute in the proceeding which outweighs the risk that the evidence may have an unfairly prejudicial effect on the defendant."
This proposal does nothing to improve our law on this point.
This sort of evidence, called similar fact evidence, should only be allowed to be used in exceptional circumstances, for the obvious reason that it will have a tendency to bias the tribunal of fact (usually the jury) against the accused.
It must be said that the current law on similar fact evidence in New Zealand is in need of clarification. This is largely due to the decision of the Court of Appeal in R v Holtz [2003] 1 NZLR 667; (2002) 20 CRNZ 14 (CA), where the criterion for admissibility was held to be whether the probative value of the similar fact evidence outweighs its illegitimately prejudicial effect:
"One of the problems with evidence of past conduct, whether labelled evidence of bad character, propensity evidence or similar fact evidence, is the tendency to look for principles of admissibility applicable to all such evidence in all circumstances. At a very general level there is the broad principle that to be admissible the evidence must be such that its probative value outweighs illegitimate prejudice to the accused in having adduced evidence of past conduct that might be given undue weight or used improperly in reasoning towards guilt of the crime charged. It is in the application of that general principle in the circumstances of particular cases that difficulties arise. Often this is because the approach adopted in one set of circumstances is taken as the rule to be applied in a quite different case. Because of the wide variation in circumstances in which the issue calls for consideration, the preferable approach in a particular case is that which determines and weighs probative value and potential prejudice in the circumstances. It is necessary to consider the nature of the evidence, what it is sought to prove, what other evidence there is and its relationship and whether the evidence relates to prior proved offending or to concurrent charges, all against the underlying dangers inherent in propensity or bad character evidence." (para 35)
The so-called weighing of probative value against prejudicial effect is problematic conceptually, and has been criticised in the leading Australian case on similar fact evidence, Pfennig v R (1995) 182 CLR 461 per McHugh J on the basis that these were "incommensurables", and the Supreme Court of Canada has agreed with his criticism, in R v Handy (2002) 164 CCC (3d) 481 (SCC); 213 DLR (4th) 385, at paras 148-149. And I have written an article on it: "Probative value, illegitimate prejudice and the accused's right to a fair trial" (2006) 29 Crim LJ 8.
In Phillips v R [2006] HCA 4 (1 March 2006) the High Court of Australia has confirmed that Pfennig continues to be the authority on similar fact evidence, binding on Australian courts. It issued a thinly-veiled rebuke to courts that sought to find reasons for departing from Pfennig, at para 60:
"It must be said at once that it is for this Court alone to determine whether one of its previous decisions is to be departed from or overruled …. Of course, in criminal cases it is often necessary for trial judges and Courts of Criminal Appeal to elaborate upon rulings of this Court; to gather together rules expressed in several cases; to apply rules to different facts; and sometimes to reconsider rules affected by later legislation. Within spaces left by the binding determinations of this Court, trial judges and intermediate courts retain their proper functions …. However, these do not extend to varying, qualifying or ignoring a rule established by a decision of this Court. Such a rule is binding on all courts and judges in the Australian Judicature."
One misinterpretation of Pfennig, alluded to briefly in Phillips at para 62, is that it is appropriate and sufficient to regard similar fact evidence as a form of circumstantial evidence. This is what had occurred in Holtz. See Adams on Criminal Law, Ch2.8.19(5), and my blog entry for 14 October 2004. A danger of that view is that it leaves the process of drawing inferences unelaborated by particular directions to the jury, as authorities on inferences tend to avoid such assistance. It also tempts the court to admit the evidence, as circumstantial evidence is just a category of admissible evidence.
Phillips and Pfennig hold that similar fact evidence must pass a high threshold before it can be admitted (para 79):
"Criminal trials in this country are ordinarily focused with high particularity upon specified offences. They are not, as such, a trial of the accused's character or propensity towards criminal conduct. That is why, in order to permit the admission of evidence relevant to several different offences, the common law requires a high threshold to be passed. The evidence must possess particular probative qualities; a strong degree of probative force; a really material bearing on the issues to be decided. That threshold was not met in this case. It was therefore necessary that the allegations, formulated in the charges brought against the appellant, be separately considered by different juries, uncontaminated by knowledge of other complaints. This is what Pfennig and other decisions of this Court require. To the extent that O'Keefe [a Queensland Court of Appeal decision critical of the High Court of Australia's judgments in Pfennig] or other authority suggests otherwise, it does not represent the law. No other outcome would be compatible with the fair trial of the appellant."
Phillips had been charged with 8 counts of sexual offending, and had unsuccessfully sought severance of those. Joinder was permitted on the basis that the evidence on each charge was admissible on the others, as evidence of similar facts. The High Court summarised the evidence and its conclusion in this way (para 56):
"The similarities relied on were not merely not "striking", they were entirely unremarkable. That a male teenager might seek sexual activity with girls about his own age with most of whom he was acquainted, and seek it consensually in the first instance, is not particularly probative. Nor is the appellant's desire for oral sex, his approaches to the complainants on social occasions and after some of them had ingested alcohol or other drugs, his engineering of opportunities for them to be alone with him, and the different degrees of violence he employed in some instances. His recklessness in persisting with this conduct near other people who might be attracted by vocal protests is also unremarkable and not uncommon."
Bad technique, indeed, as it got Phillips to court.
The trial judge, and the Queensland Court of Appeal, had reasoned that so many similar complaints could not have arisen by chance, so together they had credibility. The High Court was not so slap-dash. It identified the issue on which the challenged evidence was relevant as being consent, and it pointed out that the non-consent of one complainant could not possibly prove that another had not consented.
One thing this case illustrates is the way judges can lose their reasoning powers in cases of alleged sexual misconduct. There is, in other words, a grave danger of judicial bias (probably unconscious bias) in such cases. In those circumstances, clear rules, rather than discretions, are needed concerning the admissibility of evidence of other alleged misconduct. The Pfennig rule is that similar fact evidence is not admissible unless there is no reasonable view of it that is consistent with the accused’s innocence. It must have strong probative force, a really material bearing on the issues to be decided; its probative force must clearly transcend its merely prejudicial effect; it must have a sufficient nexus, some specific connection, with the issues; its probative force must be such as to make it just to admit it (Phillips, para 54). None of those criteria were met in Phillips (para 55).
It is sad to note that in New Zealand, the Evidence Bill proposes an approach to similar fact evidence (or, in its terminology, "propensity evidence") that applies the probative value/prejudicial effect weighing, criticism of which is noted above. Clause 39(1) of the Bill provides:
"The prosecution may offer propensity evidence about a defendant in a criminal proceeding only if the evidence has a probative value in relation to an issue in dispute in the proceeding which outweighs the risk that the evidence may have an unfairly prejudicial effect on the defendant."
This proposal does nothing to improve our law on this point.
Monday, February 20, 2006
Overcoming the void
As everyone knows, the law of Scotland has its origins in Roman-Dutch law. Its technical language can look a little strange to English jurists. A recent Privy Council decision on Scottish law illustrates how light can be shed on common issues. Here, in Ruddy and others v Procurator Fiscal [2006] UKPC D2 (6 February 2006), the Judicial Committee leant heavily on the expertise of Lord Rodger, all members agreeing with his opinion.
I will describe the case, insofar as I can, in English.
If a judicial proceeding is fundamentally flawed, so that in oft-used terminology it may be called "null and void", or "a nullity", may its result nevertheless be regarded as final?
These appeals concerned convictions and sentences that had been imposed by a tribunal that had no jurisdiction to act. They therefore, it was agreed, were in breach of the appellants’ rights to a "hearing" under s 6 of the ECHR. They had delayed 2 years before complaining of that by bringing the actions that resulted in these appeals.
Two concepts that operate in the context of errors are waiver and acquiescence. Waiver is active, and acquiescence is passive, acceptance that the error does not matter. They can operate where there is delay in seeking a remedy. In the absence of an applicable time limit on seeking redress, action must be taken within a "reasonable" time, as policy favours finality and stability. Acquiescence may be inferred from delay and the circumstances of the case (see paras 28, 37, 38, 46).
Here, if objection had been taken promptly, re-hearings could have been ordered before properly constituted tribunals and proper convictions would, inevitably, have been obtained (para 41). If a remedy was now allowed, justice might not be done in view of the effect of the passage of time.
Lord Rodger pointed to examples where errors causing the proceedings to be a "nullity" were nevertheless cured by acquiescence (para 48). Central to the acceptability of this is reasoning along these lines: a defect can be waived before the conclusion of the proceeding, so there is no reason why acquiescence after it cannot be effective (para 50). Lord Carswell elaborated this at para 58, "…in either case … [the defendant] is representing, expressly or tacitly, that the court is acceptable to him …".
In essence, a flaw in the proceedings can be waived or acquiesced in, even if it is fundamental.
This is potentially relevant to a proceeding which the Supreme Court of New Zealand will hear this week: Hansen v R (previously noted in these blogs on 19 September 2005). If the Court decides that the reverse onus provision in question, s 6(6) of the Misuse of Drugs Act 1975, puts merely an evidential, not a legal, burden of proof on the defence, questions may arise about convictions that have been incurred in trials where juries have been directed differently. While a misdirection on the standard of proof will normally result in the quashing of a conviction, it is, arguably, not so fundamental an objection as lack of jurisdiction by the tribunal. Nevertheless, given that it was possible to anticipate that the New Zealand law on reverse onuses would be brought into line with English law, can it now be said that people who were convicted before the change (assuming, of course, that Hansen does change the law) acquiesced in the error?
A straightforward solution to this would be to say that Hansen changes (again, assuming that it will change) the law from the date it is decided, and that earlier trials (but not Hansen’s, as these are still his proceedings) where heard under the law as it was then correctly understood to be.
I will describe the case, insofar as I can, in English.
If a judicial proceeding is fundamentally flawed, so that in oft-used terminology it may be called "null and void", or "a nullity", may its result nevertheless be regarded as final?
These appeals concerned convictions and sentences that had been imposed by a tribunal that had no jurisdiction to act. They therefore, it was agreed, were in breach of the appellants’ rights to a "hearing" under s 6 of the ECHR. They had delayed 2 years before complaining of that by bringing the actions that resulted in these appeals.
Two concepts that operate in the context of errors are waiver and acquiescence. Waiver is active, and acquiescence is passive, acceptance that the error does not matter. They can operate where there is delay in seeking a remedy. In the absence of an applicable time limit on seeking redress, action must be taken within a "reasonable" time, as policy favours finality and stability. Acquiescence may be inferred from delay and the circumstances of the case (see paras 28, 37, 38, 46).
Here, if objection had been taken promptly, re-hearings could have been ordered before properly constituted tribunals and proper convictions would, inevitably, have been obtained (para 41). If a remedy was now allowed, justice might not be done in view of the effect of the passage of time.
Lord Rodger pointed to examples where errors causing the proceedings to be a "nullity" were nevertheless cured by acquiescence (para 48). Central to the acceptability of this is reasoning along these lines: a defect can be waived before the conclusion of the proceeding, so there is no reason why acquiescence after it cannot be effective (para 50). Lord Carswell elaborated this at para 58, "…in either case … [the defendant] is representing, expressly or tacitly, that the court is acceptable to him …".
In essence, a flaw in the proceedings can be waived or acquiesced in, even if it is fundamental.
This is potentially relevant to a proceeding which the Supreme Court of New Zealand will hear this week: Hansen v R (previously noted in these blogs on 19 September 2005). If the Court decides that the reverse onus provision in question, s 6(6) of the Misuse of Drugs Act 1975, puts merely an evidential, not a legal, burden of proof on the defence, questions may arise about convictions that have been incurred in trials where juries have been directed differently. While a misdirection on the standard of proof will normally result in the quashing of a conviction, it is, arguably, not so fundamental an objection as lack of jurisdiction by the tribunal. Nevertheless, given that it was possible to anticipate that the New Zealand law on reverse onuses would be brought into line with English law, can it now be said that people who were convicted before the change (assuming, of course, that Hansen does change the law) acquiesced in the error?
A straightforward solution to this would be to say that Hansen changes (again, assuming that it will change) the law from the date it is decided, and that earlier trials (but not Hansen’s, as these are still his proceedings) where heard under the law as it was then correctly understood to be.
Friday, February 17, 2006
The "overarching requirement" of fairness
The interrelationship between the requirements that statements be made voluntarily, and that trials be fair, was central to the unanimous decision of the Judicial Committee in Shabadine Peart v R (Jamaica) [2006] UKPC 5 (14 February 2006).
Which is the dominant requirement? If a confession was made voluntarily, should it always be admissible? If it was obtained wrongfully, but still given voluntarily, should it always be admissible? Obtaining a statement wrongfully can result in lies being told which unfairly diminish the credibility of the defence. This was the position in Shabadine Peart. Should such a statement be admissible, because it was made voluntarily, even though it could result in unfairness at trial? Is it necessary that the effect of the misconduct in obtaining the statement would be to make the trial unfair, or can exclusion of the wrongfully obtained statement be justified on broader grounds of public policy?
I should note, at this point, that this case focuses on the effect of admission of the wrongfully obtained statement on trial fairness, when that matter would not arise unless the statement should have been excluded on public policy grounds. If there were no public policy grounds to exclude the statement, there could be no trial fairness objection to its admission. Yet, once there were sufficient public policy grounds to exclude it, the appellate court had to deal with the fair trial implications of the error.
Well, aside from that matter, what does Shabadine Peart decide about trial fairness? It follows a line of cases, not cited in the judgment but referred to frequently in these blogs, that puts fairness as the overarching criterion (para 23). The most important facet of fairness is the voluntariness of the statement by the accused, and also relevant are factors such as his youth, and whether he received legal advice before making the statement.
The case involved breaches of Rule 3(b) of the Judges’ Rules, which concerns restrictions on questions that may be put to the accused after he has been arrested. The Privy Council doubted that in the circumstances the statement could be regarded as voluntary, but, even if voluntary, it was unfair to admit the statement in evidence (para 29). This was because it contradicted some of the evidence given or called by the accused at trial, and could therefore be used by the prosecution to diminish the credibility of the defence.
Well, it would only be wrong to admit such a statement if it had been wrongfully obtained in such a way as to give rise to public policy concerns over disrepute to the administration of justice. In my view, as stated above, the Board's comments on trial fairness were unnecessary and misleading.
At this point in the judgment it was necessary to consider the application of the proviso – itself an interesting topic, considered in these blogs. It was held that if the statement had been excluded, the course of the trial may have been very different and the defence may have succeeded (para 30). This is recognisable as the criterion of whether the error at trial had resulted in the loss of a real chance of acquittal. That was sufficient to prevent application of the proviso. It should be noted that the Board is not, here, relying on the trial fairness aspect of application of the proviso, but instead is, properly, determining whether the error at trial had had significance for its result. Inconsequential errors in the application of the public policy discretion do not, by definition, matter.
Some aspects of Rule 3(b) are also mentioned. These include the history and rationale for the Rule, the distinction between the suspicion and accusation phases of the police inquiry, and judicial experience of the tendency of those who have told the truth to tell lies to avoid pressure (paras 18 – 20).
Which is the dominant requirement? If a confession was made voluntarily, should it always be admissible? If it was obtained wrongfully, but still given voluntarily, should it always be admissible? Obtaining a statement wrongfully can result in lies being told which unfairly diminish the credibility of the defence. This was the position in Shabadine Peart. Should such a statement be admissible, because it was made voluntarily, even though it could result in unfairness at trial? Is it necessary that the effect of the misconduct in obtaining the statement would be to make the trial unfair, or can exclusion of the wrongfully obtained statement be justified on broader grounds of public policy?
I should note, at this point, that this case focuses on the effect of admission of the wrongfully obtained statement on trial fairness, when that matter would not arise unless the statement should have been excluded on public policy grounds. If there were no public policy grounds to exclude the statement, there could be no trial fairness objection to its admission. Yet, once there were sufficient public policy grounds to exclude it, the appellate court had to deal with the fair trial implications of the error.
Well, aside from that matter, what does Shabadine Peart decide about trial fairness? It follows a line of cases, not cited in the judgment but referred to frequently in these blogs, that puts fairness as the overarching criterion (para 23). The most important facet of fairness is the voluntariness of the statement by the accused, and also relevant are factors such as his youth, and whether he received legal advice before making the statement.
The case involved breaches of Rule 3(b) of the Judges’ Rules, which concerns restrictions on questions that may be put to the accused after he has been arrested. The Privy Council doubted that in the circumstances the statement could be regarded as voluntary, but, even if voluntary, it was unfair to admit the statement in evidence (para 29). This was because it contradicted some of the evidence given or called by the accused at trial, and could therefore be used by the prosecution to diminish the credibility of the defence.
Well, it would only be wrong to admit such a statement if it had been wrongfully obtained in such a way as to give rise to public policy concerns over disrepute to the administration of justice. In my view, as stated above, the Board's comments on trial fairness were unnecessary and misleading.
At this point in the judgment it was necessary to consider the application of the proviso – itself an interesting topic, considered in these blogs. It was held that if the statement had been excluded, the course of the trial may have been very different and the defence may have succeeded (para 30). This is recognisable as the criterion of whether the error at trial had resulted in the loss of a real chance of acquittal. That was sufficient to prevent application of the proviso. It should be noted that the Board is not, here, relying on the trial fairness aspect of application of the proviso, but instead is, properly, determining whether the error at trial had had significance for its result. Inconsequential errors in the application of the public policy discretion do not, by definition, matter.
Some aspects of Rule 3(b) are also mentioned. These include the history and rationale for the Rule, the distinction between the suspicion and accusation phases of the police inquiry, and judicial experience of the tendency of those who have told the truth to tell lies to avoid pressure (paras 18 – 20).
Wednesday, February 15, 2006
Easy does it!
The Supreme Court’s decision in Thompson v R [2006] SCNZ 3 (15 February 2006), in which answers to questions at trial were held not to have caused unfairness, gives us an opportunity to reflect on the art of cross-examination.
In my view, the following ten are fundamental points that must be remembered by counsel faced with the task of cross-examining a witness:
1. Find out beforehand the worst that the witness could say, from your client’s point of view.
2. It is unlikely that the witness will be compelled, by your questioning, to tell lies that suit your client.
3. If you are trying to get the witness to change an assertion, it may be best to start by suggesting reasons the witness may be honestly mistaken. This is because an aggressive approach to cross-examination is less likely to find sympathy with the jury than an approach which assumes the witness will in good faith acknowledge ordinary human failings.
4. An aggressive approach is unlikely to succeed with an aggressive witness. Gently hinting at common sense reasons for the witness’s error is preferable, especially if the jurors are likely to accept them.
5. Where the witness can be contradicted by his own previous statement, make the most of that by getting him to give as much "detail" as possible first, before putting the contradicting statement. This applies as long as the witness is unlikely to have been wrong in the previous statement, and where that earlier statement is more favourable to your client than the witness’s present testimony.
6. There is a duty to put your client’s case to the witness. This can be done point by point, in questions that take the form "Didn’t X?", "Didn’t Y?" etc. This is stronger than the formula that is commonly encountered: "My client will give evidence and say that X, what do you say to that?"
7. Don’t ask unnecessary questions. Don’t ask questions that are open-ended, such as "Why do you say that?" Lead the witness as much as possible, to give your questioning a focus that the jury understands. One of the purposes of cross-examination is to get the witness to say what you want him to say, to score a point and/or to undermine your opponent’s case.
8. Don’t ask questions that you don’t know the answer to, unless the whole tenor of the case points in the direction of the answer you want and any other answer would lack credibility.
9. If you have to attack the witness as a liar, remember that this strategy may result in the jury learning of your client’s own bad character (including previous convictions).
10. If your client is of good character, and if you rely on that as part of your case, take opportunities to use it in cross-examination; this can be tactically advantageous as it should result in the judge directing the jury on your client’s good character.
In Thompson, experienced trial counsel was drawn in to engaging in what the Court of Appeal called a "slugfest" with the witness: see R v Thompson 16/6/05, CA445/04, para 56. This, of itself, was not necessarily an error, as the Court of Appeal noted, at para 66:
"… robust cross-examination is one of the many options open to counsel, who must be accorded wide discretion."
But the Court added, para 69:
"Generally, counsel find it politic to conduct cross-examination with ‘a measure of courtesy’ …."
The conclusion, upheld by the Supreme Court, was that the evidence that was inadmissible but which was given in cross-examination, when seen in context, and in the light of a strong warning by the judge to the jury about its use, did not cause a miscarriage of justice.
In my view, the following ten are fundamental points that must be remembered by counsel faced with the task of cross-examining a witness:
1. Find out beforehand the worst that the witness could say, from your client’s point of view.
2. It is unlikely that the witness will be compelled, by your questioning, to tell lies that suit your client.
3. If you are trying to get the witness to change an assertion, it may be best to start by suggesting reasons the witness may be honestly mistaken. This is because an aggressive approach to cross-examination is less likely to find sympathy with the jury than an approach which assumes the witness will in good faith acknowledge ordinary human failings.
4. An aggressive approach is unlikely to succeed with an aggressive witness. Gently hinting at common sense reasons for the witness’s error is preferable, especially if the jurors are likely to accept them.
5. Where the witness can be contradicted by his own previous statement, make the most of that by getting him to give as much "detail" as possible first, before putting the contradicting statement. This applies as long as the witness is unlikely to have been wrong in the previous statement, and where that earlier statement is more favourable to your client than the witness’s present testimony.
6. There is a duty to put your client’s case to the witness. This can be done point by point, in questions that take the form "Didn’t X?", "Didn’t Y?" etc. This is stronger than the formula that is commonly encountered: "My client will give evidence and say that X, what do you say to that?"
7. Don’t ask unnecessary questions. Don’t ask questions that are open-ended, such as "Why do you say that?" Lead the witness as much as possible, to give your questioning a focus that the jury understands. One of the purposes of cross-examination is to get the witness to say what you want him to say, to score a point and/or to undermine your opponent’s case.
8. Don’t ask questions that you don’t know the answer to, unless the whole tenor of the case points in the direction of the answer you want and any other answer would lack credibility.
9. If you have to attack the witness as a liar, remember that this strategy may result in the jury learning of your client’s own bad character (including previous convictions).
10. If your client is of good character, and if you rely on that as part of your case, take opportunities to use it in cross-examination; this can be tactically advantageous as it should result in the judge directing the jury on your client’s good character.
In Thompson, experienced trial counsel was drawn in to engaging in what the Court of Appeal called a "slugfest" with the witness: see R v Thompson 16/6/05, CA445/04, para 56. This, of itself, was not necessarily an error, as the Court of Appeal noted, at para 66:
"… robust cross-examination is one of the many options open to counsel, who must be accorded wide discretion."
But the Court added, para 69:
"Generally, counsel find it politic to conduct cross-examination with ‘a measure of courtesy’ …."
The conclusion, upheld by the Supreme Court, was that the evidence that was inadmissible but which was given in cross-examination, when seen in context, and in the light of a strong warning by the judge to the jury about its use, did not cause a miscarriage of justice.
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.
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.
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.
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:
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.
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.
"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.
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.
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.
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.
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.
Subscribe to:
Posts (Atom)