Public policy exclusion of evidence is sometimes still spoken of by courts as if it was a means of protecting the accused’s right to a fair trial. This was done yesterday by the Privy Council in Williams v R (Jamaica) [2006] UKPC 21 (25 April 2006). A statement had been obtained from the accused when he was aged 12, in breach of the procedures laid down for the interviewing of young suspects (para 27). The Board concluded, at para 28:
"…the circumstances of the appellant's detention and of the taking of the statement were such as to create a significant amount of unfairness to him. Their Lordships cannot conclude that in all the circumstances of the case it was fair to admit the statement."
While the Board was not expressly referring to trial fairness, the expressions "unfairness to him" and it not being "fair to admit" the statement do suggest that trial fairness is the object of concern. Another point in this extract that is misleading is the expression "a significant amount of unfairness".
My view is that in these situations the courts are not excluding the evidence for reasons that have anything to do with trial fairness. The probative value of the evidence is not relevant to the decision on admissibility in these cases (see, for example, my blog entry for 4 April 2006 concerning "cogency"). It is not the accused’s guilt that matters, it is the objectionable way in which the evidence was obtained that is critical. This is why the discretion, in this area, is best called the public policy discretion. It is true that, historically, this discretion evolved from – and remains part of - the court’s inherent power to prevent an abuse of process. In turn, that power has been referred to as giving rise to a discretion to exclude evidence in the interests of "fairness", or, sometimes, "fairness to the accused". But these are not references to trial fairness.
One of the reasons for making this distinction between the public policy exclusion of evidence, and the exclusion of evidence to ensure trial fairness, is to preserve the concept of the absolute nature of the accused’s right to a fair trial. In the above quotation from para 28 of Williams, the expression "a significant amount of unfairness" could, wrongly, suggest that fairness of trial exists in gradations, and that some forms of trial unfairness are acceptable. I have given examples of misuse of this terminology in "The Duty to Prevent an Abuse of Process by Staying Criminal Proceedings" in Essays on Criminal Law – A Tribute to Professor Gerald Orchard (Brookers Ltd, 2004), 133, 146.
What, it is respectfully suggested, the Privy Council should have said in Williams, is that the breaches of the Directions on the conduct of interviews of young persons that occurred in this case were sufficiently serious that admission of the statement obtained thereby would be an abuse of process. The evidence was excluded to prevent the administration of justice being brought into disrepute, as would occur if the courts appeared to endorse the police misconduct. In that context it is appropriate, if one must use the "fairness" terminology, to speak of degrees of unfairness, because official misconduct comes in degrees.
There may be cases where trial fairness could be relevant to the admission into evidence of a statement that had been obtained wrongfully. But it must be remembered that, at the stage when the ruling on admissibility has to be made, the effects of admission on the defence will not be known: the Judge will not know whether the defendant intends to give or call evidence, let alone what any such evidence would be. There is a difference between using fairness as grounds for excluding evidence, and using fairness in considering, as an appellate court, whether to apply the proviso. It is the appellate court that is best placed to evaluate the effect of an erroneous admission of evidence, and, when the public policy discretion has been exercised in favour of admitting the evidence, it may be only retrospectively that the trial can be said to have been unfair.
Wednesday, April 26, 2006
Seeing and believing
Eyewitness identification evidence may need to be treated with some circumspection, and juries are usually given a direction on the special need for caution before relying on such evidence. In some jurisdictions, these warnings are required by statute, although, as is the case in New Zealand, the points required to be covered are not spelt out in great detail. The common law antecedent of these directions is known as the Turnbull direction, originating in the English Court of Appeal’s decision R v Turnbull [1977] QB 224.
In Edwards v R (Jamaica) [2006] UKPC 23 (25 April 2006) the Privy Council indicated that there are some practices that should not be permitted at trials where identification is an issue and the prosecution relies on evidence of an eyewitness to the offence. These are:
(1) The eyewitness should not be permitted to identify the accused in the dock as the offender. The prosecution should, in general, adopt other means for establishing that the accused is the person who was arrested (para 22):
"… it is only in the most exceptional circumstances that any form of dock identification is permissible: cf the discussion in the Scottish devolution appeal Holland v HM Advocate [2005] UKPC D1, 2005 SLT 563. It may be borne in mind that this was far from being a first identification and it can fairly be said that the dock identification may have had little impact on the minds of the jury. It is, however, an undesirable practice in general and other means should be adopted of establishing that the defendant in the dock is the man who was arrested for the offence charged."
(2) A police officer should not be permitted to give in evidence his opinion on why an ID parade was not considered to be necessary (para 23).
(3) The police should not give in evidence the fact that a warrant was obtained for the arrest of the accused, or of the information on which the police acted, as this is hearsay and potentially highly prejudicial (para 23).
(4) The police should not give in evidence the fact that a potential witness was unwilling to come forward (para 23).
(5) The police should avoid confronting the eyewitness with the suspect (para 25).
There is, at this point – para 25 – a possibly unintentional suggestion by the Board that hearsay evidence might be given to establish the link between the person described to the police as the offender, and the suspect:
"The arresting officer would have been quite capable of establishing that the appellant was the person pointed out to him by Bailey [the eyewitness] near the Mango Tree Bar, so it was unnecessary to ask Bailey to come to the station to confirm that."
This, however, should be read as referring to "establishing" in an investigatory, pre-trial, sense, and not as "establishing" in evidence at trial.
In this case, the eyewitness to the killing had been standing next to the victim, and the bullet that killed the victim had passed through the eyewitness. The offender had been trying to rob the eyewitness, who suddenly and unsuccessfully tried to grab the gun. The eyewitness was hospitalised for 4 weeks, and it was 2 months after the killing before he saw the accused near the same bar. He claimed that the accused was the offender. In his first description of the offender, given 5 days after the incident while he was in hospital, the eyewitness failed to mention a prominent birthmark on the accused’s face, he was unable to say what sort of trousers the offender was wearing, and he claimed that the time he had to observe the offender was a couple of minutes although it must have been shorter than that. The circumstances in which the offender was observed were good: inside a bar in the morning with good lighting.
The Privy Council was concerned that there could have been an erroneous association of ideas arising from the location of the offence and the subsequent identification being similar, and that the judge had not adequately warned the jury of the dangers in accepting the evidence. The conviction was therefore unsafe.
In New Zealand, the Evidence Bill 2006, clause 122, almost exactly repeats the current provision on the need for judicial warning: Crimes Act 1961, s 344D. The slight difference is that instead of requiring the judge to "include the reason for the warning", the Bill requires the judge to "warn the jury that a mistaken identification can result in a serious miscarriage of justice". The need for a warning arises "In a criminal proceeding tried with a jury in which the case against the defendant depends wholly or substantially on the correctness of 1 or more visual or voice identifications of the defendant or any other person …". The inclusion of voice identification is new to the Bill.
The Bill contains other provisions relating to the admissibility of visual identification evidence. It is important to note that here the concern is with admissibility, not with the way admissible evidence is treated at trial. These provisions, for visual identification, are in clause 41, and they concern the implications of whether or not a formal identification procedure was used at the investigatory stage. The criterion for admissibility is proof, on the balance of probabilities, that the evidence is reliable. The Bill does not say to what extent, if any, this reliability should be assessed by reference to the other evidence in the case. It seems plain that the other evidence should not be included in the assessment of the reliability of the visual identification evidence, and that the focus should be on the circumstances in which the identification was made.
In Edwards v R (Jamaica) [2006] UKPC 23 (25 April 2006) the Privy Council indicated that there are some practices that should not be permitted at trials where identification is an issue and the prosecution relies on evidence of an eyewitness to the offence. These are:
(1) The eyewitness should not be permitted to identify the accused in the dock as the offender. The prosecution should, in general, adopt other means for establishing that the accused is the person who was arrested (para 22):
"… it is only in the most exceptional circumstances that any form of dock identification is permissible: cf the discussion in the Scottish devolution appeal Holland v HM Advocate [2005] UKPC D1, 2005 SLT 563. It may be borne in mind that this was far from being a first identification and it can fairly be said that the dock identification may have had little impact on the minds of the jury. It is, however, an undesirable practice in general and other means should be adopted of establishing that the defendant in the dock is the man who was arrested for the offence charged."
(2) A police officer should not be permitted to give in evidence his opinion on why an ID parade was not considered to be necessary (para 23).
(3) The police should not give in evidence the fact that a warrant was obtained for the arrest of the accused, or of the information on which the police acted, as this is hearsay and potentially highly prejudicial (para 23).
(4) The police should not give in evidence the fact that a potential witness was unwilling to come forward (para 23).
(5) The police should avoid confronting the eyewitness with the suspect (para 25).
There is, at this point – para 25 – a possibly unintentional suggestion by the Board that hearsay evidence might be given to establish the link between the person described to the police as the offender, and the suspect:
"The arresting officer would have been quite capable of establishing that the appellant was the person pointed out to him by Bailey [the eyewitness] near the Mango Tree Bar, so it was unnecessary to ask Bailey to come to the station to confirm that."
This, however, should be read as referring to "establishing" in an investigatory, pre-trial, sense, and not as "establishing" in evidence at trial.
In this case, the eyewitness to the killing had been standing next to the victim, and the bullet that killed the victim had passed through the eyewitness. The offender had been trying to rob the eyewitness, who suddenly and unsuccessfully tried to grab the gun. The eyewitness was hospitalised for 4 weeks, and it was 2 months after the killing before he saw the accused near the same bar. He claimed that the accused was the offender. In his first description of the offender, given 5 days after the incident while he was in hospital, the eyewitness failed to mention a prominent birthmark on the accused’s face, he was unable to say what sort of trousers the offender was wearing, and he claimed that the time he had to observe the offender was a couple of minutes although it must have been shorter than that. The circumstances in which the offender was observed were good: inside a bar in the morning with good lighting.
The Privy Council was concerned that there could have been an erroneous association of ideas arising from the location of the offence and the subsequent identification being similar, and that the judge had not adequately warned the jury of the dangers in accepting the evidence. The conviction was therefore unsafe.
In New Zealand, the Evidence Bill 2006, clause 122, almost exactly repeats the current provision on the need for judicial warning: Crimes Act 1961, s 344D. The slight difference is that instead of requiring the judge to "include the reason for the warning", the Bill requires the judge to "warn the jury that a mistaken identification can result in a serious miscarriage of justice". The need for a warning arises "In a criminal proceeding tried with a jury in which the case against the defendant depends wholly or substantially on the correctness of 1 or more visual or voice identifications of the defendant or any other person …". The inclusion of voice identification is new to the Bill.
The Bill contains other provisions relating to the admissibility of visual identification evidence. It is important to note that here the concern is with admissibility, not with the way admissible evidence is treated at trial. These provisions, for visual identification, are in clause 41, and they concern the implications of whether or not a formal identification procedure was used at the investigatory stage. The criterion for admissibility is proof, on the balance of probabilities, that the evidence is reliable. The Bill does not say to what extent, if any, this reliability should be assessed by reference to the other evidence in the case. It seems plain that the other evidence should not be included in the assessment of the reliability of the visual identification evidence, and that the focus should be on the circumstances in which the identification was made.
Tuesday, April 04, 2006
Balancing "cogency" of wrongfully obtained evidence
Simmons v R (Bahamas) [2006] UKPC 19 (3 April 2006) gives us an opportunity to highlight the distinction between "fairness" when that term is used in the context of the exercise of the public policy discretion to exclude evidence, and "fairness" in the separate sense of trial fairness for the accused.
The two appellants were convicted of murder. Included in the evidence against them were statements they had made to the police. These statements, referred to as confessions, although they were partly exculpatory (and, one accused who gave evidence adopted what he had said to the police), were obtained in breach of the appellants’ constitutional right to be informed of the availability of legal advice before they spoke to the police.
Breach of that right gives rise to a judicial balancing exercise to determine whether to admit or exclude the statement. This balancing exercise is a public policy discretion, having nothing to do with the fairness of the trial. It arises because of the conflict between, on the one hand, the public’s right to have suspects prosecuted and offenders brought to justice, and, on the other hand, the public’s right to have officials comply with the law in the investigation of offences. The only sanction that courts can impose on officials who act in breach of the law in this context is to exclude evidence that they obtain thereby.
Sometimes, in carrying out this balancing exercise, judges say that the cogency of the evidence is a factor strongly favouring its admission. An important observation on this point was made by the Privy Council in this case, para 26:
"The Board has one other concern about the judge's balancing of the respective interests of the prosecution and the defence on the issue of fairness: the evident importance which she attached to the confession being "very cogent evidence against Simmons." Their Lordships cannot accept that the potency of such evidence is necessarily a factor in favour of its admission. If, by denying a suspect his constitutional right to see a lawyer and perhaps be advised against making a statement, the state's case is thereby strengthened by a confession, it is by no means self-evident that fairness demands its admission rather than its exclusion."
However, in New Zealand the cogency of the evidence is routinely taken into account in this balancing exercise: R v Shaheed [2002] 2 NZLR 377 (CA), especially at paras 151-152. At para 151 the joint judgment (Richardson P, Blanchard and Tipping JJ) states: "A trial is not to be regarded as potentially unfair by reason of the admission of evidence unless that evidence may lead to an unsafe verdict." That, with respect, needs to be read with some caution. A safe verdict is not a cure-all for trial unfairness. That point was made strongly by Lord Steyn (Sir Swinton Thomas concurring) in Ebanks v R (Cayman Is) [2006] UKPC 16 (27 March 2006), blogged 28.3.06, at para 40. The Supreme Court acknowledged the same point in Sungsuwan v R [2005] NZSC 57 (25 August 2005), blogged 26.8.05, per Elias CJ at para 6 (putting as alternatives trial unfairness and unsafe verdicts), Tipping J at 112 (lack of a fair trial is itself a miscarriage of justice without the need to consider its effect on the verdict).
If I may, I should add that "cogency", which means being convincing or compelling, is always a matter for the jury. It is usually called the "weight" that is to be given to the evidence. Weight is separate from the question of admissibility, except on occasions where it is possible for a judge to conclude that no reasonable jury could give the evidence any weight. The Privy Council is correct to see no reason to link cogency to admissibility. However, this is not the end of the matter. What is being considered is not the admissibility of the evidence, but rather whether, as admissible evidence, it should be excluded because of the objectionable way in which it was obtained. The question whether convincing or compelling admissible evidence should be excluded is, appropriately, part of the weighing of the public interest in bringing suspects to justice. There is, though, a difficulty: the cogency of the evidence is also appropriately considered on the other side of the balance, where weight has to be given to the public interest in prevention of such abuse of process as would bring disrepute to the administration of justice. One might properly object to the inclusion of "cogency" in the weighing process on the basis that it falls on both sides of the scales.
That aside, having said that the exclusion of evidence to prevent the trial being unfair to the accused is separate from the public policy discretion, I should now make the distinction clear. A fair trial for the accused is one where the law is accurately applied and the facts are determined without bias. There may be flow-on effects of a wrongful exercise of the public policy discretion, in the sense that the trial may not be one where the law has been accurately applied. The question, in terms of trial fairness, of the significance of the error, will be determined, not by the strength of the other evidence against the accused, but by whether the error put the accused at a disadvantage in the trial. I made this point in the blog entry on 29 March 2006, discussing Gilbert v R (Grenada) [2006] UKPC 15 (27 March 2006).
Simmons is an example of the error in applying the public policy discretion not affecting the fairness of the trial. The statements in issue were partly exculpatory and the accused who gave evidence adopted what he had said. The judgment does, however, focus on the strength of the other evidence of guilt, concluding, para 31, that acquittals would have defied all reason. The relevance of this point is that the proviso could be applied: the error in admitting the statements did not amount to a "substantial miscarriage of justice." The error caused neither trial unfairness, nor the loss of a real chance of acquittal.
The two appellants were convicted of murder. Included in the evidence against them were statements they had made to the police. These statements, referred to as confessions, although they were partly exculpatory (and, one accused who gave evidence adopted what he had said to the police), were obtained in breach of the appellants’ constitutional right to be informed of the availability of legal advice before they spoke to the police.
Breach of that right gives rise to a judicial balancing exercise to determine whether to admit or exclude the statement. This balancing exercise is a public policy discretion, having nothing to do with the fairness of the trial. It arises because of the conflict between, on the one hand, the public’s right to have suspects prosecuted and offenders brought to justice, and, on the other hand, the public’s right to have officials comply with the law in the investigation of offences. The only sanction that courts can impose on officials who act in breach of the law in this context is to exclude evidence that they obtain thereby.
Sometimes, in carrying out this balancing exercise, judges say that the cogency of the evidence is a factor strongly favouring its admission. An important observation on this point was made by the Privy Council in this case, para 26:
"The Board has one other concern about the judge's balancing of the respective interests of the prosecution and the defence on the issue of fairness: the evident importance which she attached to the confession being "very cogent evidence against Simmons." Their Lordships cannot accept that the potency of such evidence is necessarily a factor in favour of its admission. If, by denying a suspect his constitutional right to see a lawyer and perhaps be advised against making a statement, the state's case is thereby strengthened by a confession, it is by no means self-evident that fairness demands its admission rather than its exclusion."
However, in New Zealand the cogency of the evidence is routinely taken into account in this balancing exercise: R v Shaheed [2002] 2 NZLR 377 (CA), especially at paras 151-152. At para 151 the joint judgment (Richardson P, Blanchard and Tipping JJ) states: "A trial is not to be regarded as potentially unfair by reason of the admission of evidence unless that evidence may lead to an unsafe verdict." That, with respect, needs to be read with some caution. A safe verdict is not a cure-all for trial unfairness. That point was made strongly by Lord Steyn (Sir Swinton Thomas concurring) in Ebanks v R (Cayman Is) [2006] UKPC 16 (27 March 2006), blogged 28.3.06, at para 40. The Supreme Court acknowledged the same point in Sungsuwan v R [2005] NZSC 57 (25 August 2005), blogged 26.8.05, per Elias CJ at para 6 (putting as alternatives trial unfairness and unsafe verdicts), Tipping J at 112 (lack of a fair trial is itself a miscarriage of justice without the need to consider its effect on the verdict).
If I may, I should add that "cogency", which means being convincing or compelling, is always a matter for the jury. It is usually called the "weight" that is to be given to the evidence. Weight is separate from the question of admissibility, except on occasions where it is possible for a judge to conclude that no reasonable jury could give the evidence any weight. The Privy Council is correct to see no reason to link cogency to admissibility. However, this is not the end of the matter. What is being considered is not the admissibility of the evidence, but rather whether, as admissible evidence, it should be excluded because of the objectionable way in which it was obtained. The question whether convincing or compelling admissible evidence should be excluded is, appropriately, part of the weighing of the public interest in bringing suspects to justice. There is, though, a difficulty: the cogency of the evidence is also appropriately considered on the other side of the balance, where weight has to be given to the public interest in prevention of such abuse of process as would bring disrepute to the administration of justice. One might properly object to the inclusion of "cogency" in the weighing process on the basis that it falls on both sides of the scales.
That aside, having said that the exclusion of evidence to prevent the trial being unfair to the accused is separate from the public policy discretion, I should now make the distinction clear. A fair trial for the accused is one where the law is accurately applied and the facts are determined without bias. There may be flow-on effects of a wrongful exercise of the public policy discretion, in the sense that the trial may not be one where the law has been accurately applied. The question, in terms of trial fairness, of the significance of the error, will be determined, not by the strength of the other evidence against the accused, but by whether the error put the accused at a disadvantage in the trial. I made this point in the blog entry on 29 March 2006, discussing Gilbert v R (Grenada) [2006] UKPC 15 (27 March 2006).
Simmons is an example of the error in applying the public policy discretion not affecting the fairness of the trial. The statements in issue were partly exculpatory and the accused who gave evidence adopted what he had said. The judgment does, however, focus on the strength of the other evidence of guilt, concluding, para 31, that acquittals would have defied all reason. The relevance of this point is that the proviso could be applied: the error in admitting the statements did not amount to a "substantial miscarriage of justice." The error caused neither trial unfairness, nor the loss of a real chance of acquittal.
Monday, April 03, 2006
You may not get what you want ...
The crime of attempting to commit a crime has come under scrutiny in L v R [2006] NZSC 18 (30 March 2006), which concerned the state of mind required for liability.
The facts of L were extraordinary. The charge, faced by L, a 49 year old woman, was attempting to sexually violate a 15 year old male. She tried to cause his (presumably erect) penis to penetrate her vagina, being reckless as to whether or not he consented to that. In the circumstances, it seems to have been the case that his lack of consent should have been apparent to her.
The legal question was whether, to be liable for the attempt, L had to intend that he not consent, or was it sufficient that she merely be reckless (ie that she knowingly took a risk) as to his lack of consent. The question arose because the crime of attempting to commit a crime requires the accused to have an intent to commit that crime.
Plainly, L intended that the penis enter her own vagina, and she did acts in an effort to bring that about. Was that intent sufficient? Or, in addition, did she need to be reckless about the male’s lack of consent? Or, did she need to intend both penetration and his lack of consent?
There has been some division among the jurists on this general issue. As Simester and Brookbanks, Principles of Criminal Law, point out at para 6.1.1, the decision is one of public policy. Criminalising attempts is often justified because of the danger posed to society by a person who sets out to cause harm. That is, one whose intention is dangerous, albeit that his (here, her) acts may not constitute actual harm. This sort of deliberation suggests that intention should be given a narrow meaning, so as to exclude people who are reckless about circumstances such as consent. But, that narrow meaning would create anomalies (as has been pointed out by Professor Ashworth, cited ibid): imagine two men who set out to have sexual intercourse with two women. Both men are reckless as to consent, and neither woman consents. One man succeeds in penetration, and is guilty of rape. The other fails to penetrate, and, under this narrow view, is guilty of neither rape nor attempt to rape.
In L the Supreme Court held that recklessness (in the sense of absence of a reasonable belief in consent) as to circumstances is sufficient for liability for the attempt: what has to be intended here is the act of penetration.
All well and good. We are left with a lingering image of the facts of the case. In future, will people charged with sexual violation claim they did not consent, and that the other person should be charged?
The facts of L were extraordinary. The charge, faced by L, a 49 year old woman, was attempting to sexually violate a 15 year old male. She tried to cause his (presumably erect) penis to penetrate her vagina, being reckless as to whether or not he consented to that. In the circumstances, it seems to have been the case that his lack of consent should have been apparent to her.
The legal question was whether, to be liable for the attempt, L had to intend that he not consent, or was it sufficient that she merely be reckless (ie that she knowingly took a risk) as to his lack of consent. The question arose because the crime of attempting to commit a crime requires the accused to have an intent to commit that crime.
Plainly, L intended that the penis enter her own vagina, and she did acts in an effort to bring that about. Was that intent sufficient? Or, in addition, did she need to be reckless about the male’s lack of consent? Or, did she need to intend both penetration and his lack of consent?
There has been some division among the jurists on this general issue. As Simester and Brookbanks, Principles of Criminal Law, point out at para 6.1.1, the decision is one of public policy. Criminalising attempts is often justified because of the danger posed to society by a person who sets out to cause harm. That is, one whose intention is dangerous, albeit that his (here, her) acts may not constitute actual harm. This sort of deliberation suggests that intention should be given a narrow meaning, so as to exclude people who are reckless about circumstances such as consent. But, that narrow meaning would create anomalies (as has been pointed out by Professor Ashworth, cited ibid): imagine two men who set out to have sexual intercourse with two women. Both men are reckless as to consent, and neither woman consents. One man succeeds in penetration, and is guilty of rape. The other fails to penetrate, and, under this narrow view, is guilty of neither rape nor attempt to rape.
In L the Supreme Court held that recklessness (in the sense of absence of a reasonable belief in consent) as to circumstances is sufficient for liability for the attempt: what has to be intended here is the act of penetration.
All well and good. We are left with a lingering image of the facts of the case. In future, will people charged with sexual violation claim they did not consent, and that the other person should be charged?
Subscribe to:
Posts (Atom)