Bias and expertise

At common law the absence of bias includes the absence of the appearance of bias. Consequently, a tribunal may be impugned for bias without actual bias needing to be proved, and its decision quashed without any aspersion being cast.

This happened in R v Secretary of State for the Home Department, ex parte Al-Hasan [2005] UKHL 13 (16 February 2005). Proceedings were quashed because the common law test was satisfied (para 37); in summary, this amounts to asking whether a fair-minded observer would think there was a real possibility that the tribunal had been predisposed, that it lacked impartiality, that it may have been improperly influenced. An honest doubt about this, entertained by the fair-minded observer, is sufficient, even if the doubt is unfounded.

In that case a prison official (deputy governor) presided over a disciplinary hearing in which the validity of an order promulgated by the prison governor was challenged. When the order had been formulated, the official had been present and had implicitly agreed to it. It was therefore unlikely that he would hold that the order was invalid. The House of Lords stressed that there was no suggestion that in presiding over the hearing he had intentionally acted wrongly.

Baroness Hale, para 16, concluded

“…Giving the order and deciding upon its lawfulness could have been more clearly separated.
… however professional the Deputy Governor was in his approach to his task, a fair-minded and informed observer would conclude that there was a real possibility that he would be pre-disposed to uphold the legality of the order.”

Plainly, a line had to be drawn between this unacceptable overlap of functions (participating in promulgating a rule and subsequently deciding on its lawfulness), and the position of judges, who often have extra-judicial experience involving the expressing of views on the appropriateness of legislation, or even proposing law reforms which are later enacted. Lord Rodger addressed this issue, para 9, concluding:

“The knowledge and expertise developed in these ways can only help, not hinder, their judicial work.”

And adding, para 10:

“Judges have not infrequently been party to decisions overruling their own previous decisions.”

As an example, Lord Rodger referred to (also para 10):

“In In re S (Minors) (Care Order: Implementation of Care Plan) [2002] 2 AC 291, Lord Mackay of Clashfern took part in a decision in which the House struck down a system adopted by a local authority for “starring” the essential milestones of their care plan adopted under the Children Act 1989. The appeal turned on identifying a cardinal principle of the Act – a piece of legislation for which Lord Mackay, as Lord Chancellor, had been the lead minister when the Bill was going through this House in its legislative capacity. More than that, as he explained, at p 327, para 108, he had actually given a lecture in which he suggested the idea of starring stages. At the beginning of the appeal, however, he informed counsel of this and they did not object to his sitting. So any question of apparent bias was resolved. Again, since Lord Mackay agreed with the decision to disapprove the starring system, the informed and fair-minded observer would have seen that he was well able to judge the matter independently and impartially when called upon to do so.”

Thus Lord Rodger recognises that judges can be vulnerable to giving the fair-minded observer an impression of bias, and the way of overcoming that is to inform counsel of potentially embarrassing matters at the beginning of the hearing and to invite objection to their continued sitting.

Lord Brown, delivering the leading speech in Al-Hasan, referred to the common law test for bias (para 30):

“The common law test for bias has been authoritatively settled by the recent decisions of this House in Porter v Magill [2002] 2 AC 357 and Lawal v Northern Spirit Ltd [2004] 1 All ER 187:
“The question is whether the fair-minded and informed observer, having considered the relevant facts, would conclude that there was a real possibility that the tribunal was biased”.

The correct procedure was (para 41):

“… to have avoided the appearance of bias [the deputy governor] would either have had to make plain at the adjudications that he himself had actually been present when the squat search order was confirmed (rather than give the impression, as he appears to have done, that he had known nothing of it) and sought the prisoners’ consent to his nevertheless hearing the charges, or alternatively stood down to enable them to be heard by a different governor (if necessary from another prison) without any such previous involvement in the case.”

Damages and breach of the right to a fair hearing

Breach of the right to a fair trial will usually be vindicated by the declaration of the appellate court that there had been a breach, and a quashing of the conviction. There may also be an order for a retrial, which, in a remote sort of way, can also be of advantage to the accused if he obtains an outright acquittal. Awards of damages for breach of fair trial rights are rare, and when made have been “noteworthy for their modesty”: R v Secretary of State for the Home Department, ex parte Greenfield [2005] UKHL 14 (16 February 2005), per Lord Bingham at para 17, in a judgment with which the other members of the House concurred.

Special damages may be awarded where a causal connection can be shown between the breach of the right to a fair trial and a loss for which compensation is claimed (para 11), and general damages can arise in rare cases where it can be shown that the applicant has been deprived of a real chance of a better outcome (para 14), or where there has been physical or mental suffering attributable to the breach of the right (para 16).

Where damages are appropriate, should domestic courts apply domestic scales of awards, or should they follow international law precedents? In Greenfield, the position was governed by the statutory relationship between the domestic courts and the Strasbourg court. Some points considered material by Lord Bingham at para 19 of Greenfield suggest analogous questions that may usefully be asked on this issue:

  • (1) is the relevant rights legislation a torts statute?
  • (2) has the rights statute been enacted with a view to giving remedies that improve on those at common law?
  • (3) is there any legislative direction concerning the relevance of international awards of damages?

In Greenfield, it was accepted by the respondent that, in the light of a decision of the European Court that had been given after the lower courts’ decisions in this case, there had been breaches of the right to a fair hearing because the tribunal (a deputy controller determining an allegation of breach of prison rules) was not independent or impartial, and because the applicant had been denied legal representation notwithstanding that he had a lawyer who was available to act. Nevertheless, the House of Lords found that the hearing appeared to have been conducted in an exemplary way, and it would not speculate as to whether legal representation would have made a difference to the outcome. There was, furthermore, nothing special in the case that warranted special damages for anxiety or frustration. Damages were declined but declarations of the breaches of the fair trial rights were made. Costs were reserved for further submission.

Obviously, these declarations will result in improved procedures for other people charged with breaches of prison discipline. It seems, however, that Mr Greenfield had an additional 21 days of imprisonment added to his sentence in circumstances where a conviction would have been quashed if he had been subject to the jurisdiction of the criminal court. This case is a special example of the general difficulties that people who successfully appeal convictions face in obtaining a remedy for such sentences as they may have served.

Jury secrecy

To what extent, really, is the law applied in the jury room? The almost impenetrable veil of secrecy behind which jury deliberations are conducted prevents exposure of such injustice as may occur. The law deems the secrecy to be worth that price.

A glimpse into what jurors can get up to is given in R v Smith (No 2) [2005] UKHL 12 (16 February 2005), at para 11. The remedy, quashing the convictions and potentially ordering a retrial, fits the particular case, but it was only possible because one juror was concerned enough to write to the judge during the deliberations.

There is a case for observing jury deliberations, for example, by closed circuit television monitoring, to ensure no errors are made in the application of the law. Of course issues of juror anonymity and appropriate intervention would arise, as would problems of catering for unrepresented accused persons.

How much honesty can the system cope with?

Miscarriage of justice, or inconsequential error?

When can an appellate court hold that, notwithstanding an error at trial, conviction of the accused was not a miscarriage of justice?

The difficulties in deciding this issue are illustrated by the 3-2 split in the Privy Council decision Dial v The State (Trinidad and Tobago) [2005] UKPC 4 (14 February 2005). The appellants were seeking to overturn convictions in respect of which they had been sentenced to death.

I have, in the last entry on this blog, pointed out that the implications of there being no power to direct juries to convict (R v Wang [2005] UKHL 9, 10 February 2005), have some relevance to this issue. Wang emphasised the fundamental point that at trial, it is the jury that carries out the function of applying the law to the facts. But from the perspective of a court of first appeal, the view is slightly different. Where there is a second appeal, the court’s function may again be seen differently.

The majority in Dial described the approach of a court of first appeal as follows (para 42):

“…the question arising for the Appeal Court’s determination is whether [new evidence, or an error at trial] realistically places the appellant’s guilt in reasonable doubt – whether, in other words, the verdict is now to be regarded as unsafe. That necessarily must depend upon all the evidence in the case.” [emphasis in the judgment]

On second appeal, the court looks at whether the first appellate court considered the relevant matters and whether it was accordingly entitled to reach the conclusion it did. While the majority in Dial (Lords Bingham, Carswell and Brown in a joint judgment) held that the appellate court’s conclusion should be sustained, the minority (Lords Steyn and Hutton in separate judgments) disagreed. Lord Steyn (para 54) held:

“Nowhere in … the judgment, does the Chief Justice [of Trinidad and Tobago] pause to consider how the jury (or a reasonable person circumstanced as the jury were) would have viewed the potential impact of the [fresh evidence]. That was an issue which the Court of Appeal as a matter of legal principle should have squarely faced. The Court of Appeal did not do so. It is a fundamental flaw in the judgment. In these circumstances it is not appropriate to defer to the Court of Appeal as the court who heard the new evidence.”

Lord Steyn concluded with a caution for appellate courts:

“64. It is always important for a judge to bear in mind what Justice Learned Hand in his famous address during the Second World War, in Central Park, New York City, called the spirit of liberty. He said that the spirit of liberty is the spirit which is not too sure that it is right. The need for such an approach is immeasurably increased where the issue at stake is killing a man by the cruel and barbaric punishment of the death penalty. This spirit is not evident in the judgment of the Chief Justice. Instead his judgment is expressed in certainties and absolutes with some questionable and speculative assumptions in favour of the state.”

Lord Hutton, also dissenting, referred to the law on how the court of first appeal should approach the matter:

“74. In R v Pendleton, at para 19, [2002] 1 WLR 72 the House of Lords affirmed that the guiding principle is that stated by Viscount Dilhorne in Stafford v Director of Public Prosecutions [1974] AC 878 at 906:
“While … the Court of Appeal and this House may find it a convenient approach to consider what a jury might have done if they had heard the fresh evidence, the ultimate responsibility rests with them and them alone for deciding the question [whether or not the verdict is unsafe].”
“In R v Pendleton, at para 19, Lord Bingham of Cornhill stated:
“The Court of Appeal can make its assessment of the fresh evidence it has heard, but save in a clear case it is at a disadvantage in seeking to relate that evidence to the rest of the evidence which the jury heard. For these reasons it will usually be wise for the Court of Appeal, in a case of any difficulty, to test their own provisional view by asking whether the evidence, if given at the trial, might reasonably have affected the decision of the trial jury to convict. If it might, the conviction must be thought to be unsafe.”

This reflects the inevitable distance of the appellate courts from the witness box. Furthermore, as evidence is recounted and digested on appeal, errors may creep into the narrative. Lord Steyn, at para 63, points out how the majority apparently allowed this to happen in their reliance on prosecution evidence which had not, in fact, been given. “The reliance on this factor in the majority judgment is unjustified and a failure of due process.”

When the court of second appeal finds an error of reasoning in the court of first appeal, the court of second appeal carries out the exercise as it should, in its view, have been done. We can only guess what the majority in Dial would have concluded had it found an error in the court below, and had the majority not (assuming Lord Steyn’s due process criticism is correct) made a mistake about the extent of the prosecution evidence.

There are lessons to be learnt from Dial, and they include the danger of an appellate court being dogmatic about the facts of the case. An analogy that I suggest is useful is with the approach that a trial judge takes to submissions of “no case”: R v Flyger [2001] 2 NZLR 721; (2000) 18 CRNZ 624 (CA). This, of course, is only an analogy in a loose sense, calling for lateral thinking rather than direct transposition of concepts. The question on a no case submission is whether there is evidence which, if accepted, would be sufficient to support a conviction. This might be transformed into an appropriate question on appeal where miscarriage of justice is in issue: if the error had not occurred at trial (or, as the case may be, if the fresh evidence had been available), could there have been stronger grounds for acquitting the accused than there had been at the flawed trial?

"Applying" the law

A judge may not in any circumstances direct a jury to find an accused person guilty: R v Wang [2005] UKHL 9 (10 February 2005).

This is so even where the defence had the burden of raising a defence and had failed to discharge that burden, and also where the facts are agreed at trial and there is nothing calling for adjudication and there is no basis upon which the accused can properly avoid conviction on the uncontested facts. The Crown’s argument that in these circumstances a directed guilty verdict would be acceptable (para 2) was rejected in the considered opinion of the Appellate Committee in Wang.

This does not affect certain recognised areas: the power of a judge to direct an acquittal; the power of a judge to withdraw a defence from the jury’s consideration if there is no evidence whatever to support it; and the absence of any requirement for the judge to direct the jury on issues not raised by the defence. See Wang, para 3.

At root is the fundamental distinction between the functions of the judge and the jury. In particular, it is the jury that applies the law to the facts. The judge does not apply the law to the facts. This process of “applying” is critical. It amounts to a constitutional check on the powers of the executive. The Committee cited remarks by (the then) Sir Patrick Devlin, writing in 1956, who said (Hamlyn Lectures, “Trial by Jury”, pp 160, 162), that the leaving of the last word to the jury was

“an insurance that the criminal law will conform to the ordinary man’s idea of what is fair and just. If it does not, the jury will not be a party to its enforcement …. The executive knows that in dealing with the liberty of the subject it must not do anything which would seriously disturb the conscience of the average member of Parliament or of the average juryman. I know of no other real checks that exist today upon the power of the executive.”

Acknowledgement of this fundamental principle will require some revision of the standard way of directing juries. The judge usually tells the jury that if they find certain facts proved beyond reasonable doubt, then their verdict must be guilty. That is contrary to Wang. What the jury have to be told is that their task is to apply the law, the ingredients of the offence as set out by the judge, to the facts that they find proved beyond reasonable doubt in deciding whether they find the accused guilty. Further elaboration is not necessary, and in particular it would be wrong to tell the jury that they are free to find the accused not guilty even if they are sure the prosecution has proved its case.

Even where the result of applying the law to the facts would appear to be a foregone conclusion, there may nevertheless be “nuances (to adopt the language of Lord Keith in DPP v Stonehouse [1978] AC 55) not recognised by the judicial mind.” (Wang, para 17.)

This calls to mind the problem of when an appellate court may properly hold that, notwithstanding some error of law at trial, there was no miscarriage of justice because the accused was not deprived of a real chance of acquittal. A particularly intriguing example is R v Howse [2003] 3 NZLR 767 (CA). Another is R v Bain [2004] 1 NZLR 638 (CA). Both these cases are subject to Privy Council consideration.

Collateral evidence: rule or discretion?

In contrast to revising hearsay laws that have been considered by the legislature and by the Law Commission, as was done by the House of Lords in R v Hayter – see my blog entry for Monday, February 07, 2005 – the High Court of Australia has resisted an invitation to alter the law. In Nicholls v R [2005] HCA 1 (3 February 2005), a case where the circumstances would have made any such revision obiter, one of the issues involved the collateral evidence rule.

The rule that answers to questions on collateral issues cannot be undermined by cross-examination is potentially a source of injustice if insufficient recognition is given to the exception concerning challenges to the creditworthiness of the witness. How relevant is it that the witness may have been unreliable on matters collateral to the issues at trial? This is obviously an area that should be treated as discretionary, rather than one for the application of rules.

Judges often have to decide between the alternatives of excluding evidence and admitting it, perhaps subject to a warning to the jury about the use to which such evidence may be put. I have discussed this in “The Duty to Prevent an Abuse of Process by Staying Criminal Proceedings” in Robertson (ed), “Essays on Criminal Law: A Tribute to Professor Gerald Orchard” (2004, Brookers Ltd, Wellington) 133. As I pointed out there, such decisions involve the balancing of the probative value of evidence against its prejudicial effect.

Readers who seek more agony may wish to consult my “Probative value, illegitimate prejudice and the accused’s right to a fair trial” (2005) 29 Crim LJ 8, where the balancing exercise involved in this sort of decision is analysed, and it is suggested (I summarise the suggestion) that the decision comes down to admitting evidence if it has significant probative value and if trial fairness would not be compromised.

This would be a satisfactory way of approaching the admissibility of challenges to answers to questions on collateral issues. Of course we are considering something analogous to the other side of the coin because what is usually at stake when the probative/prejudicial exercise is undertaken is the admissibility of prosecution evidence. Usually the collateral evidence rule affects evidence that the defence seeks to adduce, albeit through a prosecution witness. Looking at the decision from this perspective, the first consideration would be: would the evidence be probative on the issue of the witness’s reliability on a matter in dispute in the case? The second consideration would be, would preventing the defence from seeking the evidence affect the fairness of the trial? We can easily see that these come to much the same thing. That is acceptable, because we are not balancing them against each other.

In Nicholls v R at para 53, McHugh J, who would have changed the collateral evidence rule, put it this way:
“… Where the interests of justice are likely to be advanced by admitting evidence tending to destroy the credibility of a witness, courts should hesitate to reject such evidence. Thus, where a circumstance affecting credibility is so inextricably connected with a fact in issue that it will probably determine that fact, a trial judge should generally admit evidence of that circumstance. Evidence of such a circumstance should not be excluded merely because it is not within the established exceptions to the collateral evidence rule. In Natta (1991) 32 FCR 282 at 300 the Full Federal Court concluded that a collateral matter could be pursued beyond cross-examination “in the interests of justice, whether or not it came within any of the traditional exceptions to the rule against evidence on collateral issues.”

Principled evolution, or more mental gymnastics?

Funnily enough, the same day as the High Court of Australia heard the phrase “mental gymnastics” used by one of its Judges (Kamleh v R [2005] HCA 2 (3 February 2005) per Kirby J; see my blog entry for Friday, February 04, 2005), Lord Carswell in the House of Lords used the same expression: R v Hayter [2005] UKHL 6 (3 February, 2005).

The fundamental tussle in Hayter was over whether the House of Lords should judicially change the law relating to hearsay, when both the Law Commission and Parliament had not. By a majority of 3-2 the law was changed. Lords Brown and Carswell dissented.

Two aspects of Hayter are of interest here. First, should a judge, faced with a no-case submission at the close of the prosecution case in a joint jury trial, be permitted to make a provisional ruling to the effect that there is a case against one accused if and only if the jury convicts another accused? Second, should the fact of one accused’s guilt, proved by his confession in the absence of the other accused, be able to be used as evidence against that other accused? Because the second of these questions was answered “yes”, the first question disappeared as an issue, as the confession became evidence against the other accused thus constituting the case to answer.

An important limitation to the scope of this change in the law was specified: the confession by one accused (in the absence of the other) is only evidence against the other to the extent that it was made against the self-interest of the confessor; it is not evidence against the other accused to the extent that it asserts the guilt of that other accused. This recognises the basis upon which the confession can be assumed to be reliable.

It should also be remembered that this new exception to the hearsay rule only comes into play when the confession of one accused is relevant to the guilt of another.

This is, no doubt, a sensible development. Lord Steyn, at para 25, called it a “a principled evolution in keeping with modern developments, statutory and judge made, which corrected some of the worst absurdities of the law of evidence of a bygone era.” He likened it, unconvincingly, to the joint enterprise exception to the hearsay rule. I say “unconvincingly” because the confession was not in furtherance of the pursuit of the common enterprise on the facts of Hayter, and the analogy is potentially confusing. The “in furtherance” requirement is what gives the statement its apparent reliability in joint enterprise cases, and a foundation in reliability is essential to the “principled evolution” of which Lord Steyn speaks.

The mental gymnastics that concerned Lord Carswell, at para 73, arose on the assumption that the hearsay rule remained unchanged, so that the jury would use the contents of the statement only against its maker, yet the guilt of that person, but not the contents of his confession, could be used against the other accused. However, on the law as changed by the majority, gymnastics are still required when a statement has elements of confession together with elements of accusation against the co-accused. In such cases, the confessional aspects are now admissible against the co-accused, but not (as has always been the case, except where the joint enterprise exception applies) the accusatory elements. Such gymnastics are not all that difficult.

In summary, it seems that the proposed limits of the new rule are:

  • joint trial
  • the confessing accused must be guilty for it to be possible that the other can be convicted
  • only confessional evidence against self-interest is admissible against the other accused, for example to explain circumstantial evidence

However, when one remembers that the rules against hearsay were developed to guard against unreliable evidence, the appropriateness of these limits is difficult to justify once it is accepted that the against-interest confession is reliable. There is a likelihood that Hayter will encourage a shift towards reliability as the criterion for admissibility.

Accepting the appropriateness of this development of the hearsay rule, we may nevertheless have some doubts about suggestions that Lord Steyn made about the first question referred to here: the no-case ruling. In remarks that seem to have been intended to be of general application, at para 28, he said that the dynamics of a criminal trial may require that rulings of a conditional nature be given. This is correct, to an extent, as far as the example he gave is concerned: the confronting of the accused with accusations that he does not forthrightly deny. There may be situations where the judge has to let evidence of accusations be given, subject to later evidence that he did not deny them; but this, in the context of jury trials, is a little unlikely, as the judge will have the depositions and will already know whether the accused denied such accusations. Lord Steyn thought these situations to be analogous to that of the no-case submission at the end of the prosecution case. Plainly, the situations are completely different. To suggest otherwise is to diminish the significance of the closing by the prosecution of its case.