AN CHÚIRT UACHTARACH
THE SUPREME COURT
[S:AP:IE:2021:000016]
BETWEEN:
MICHAEL MURPHY
APPELLANT
-AND-
THE DIRECTOR OF PUBLIC PROSECUTIONS
RESPONDENT
O’Donnell C.J.
Dunne J.
Charleton J.
O’Malley J.
Woulfe J.
Judgment of Ms. Justice Iseult O’Malley delivered the 9th day of November, 2021.
Introduction
1. The appellant stood trial in Waterford Circuit Court on two charges related to what is commonly referred to as “diesel laundering”. A defence application to exclude certain prosecution evidence as inadmissible, which necessitated a voir dire, was only partially successful, and an application for a direction was unsuccessful. The jury was, however, unable to reach a verdict and the matter will, therefore, be retried. The appellant does not take issue with that state of affairs - what he desires is that the trial should be conducted before a different judge. An application to the original trial judge to recuse himself has been refused.
2. These judicial review proceedings seek orders quashing that refusal and granting related reliefs. The appellant was unsuccessful in the High Court (MacGrath J. - see [2019] IEHC 918) and the Court of Appeal (judgment delivered by the President - see [2020] IECA 334).
3. In essence, the appellant’s case is that because the trial judge made findings relating to the credibility of prosecution witnesses in the course of the voir dire, the question of objective bias arises. He contends that an informed and reasonable bystander would conclude that he would not receive a fair or impartial retrial, because the judge will be affected by his memories and perceptions of the first trial, and that the judge should therefore have acceded to a request to recuse himself.
4. To put the debate in context, it should be noted that this is not an issue that, in practice, often gives rise to dispute. The parties have informed the Court that trial judges in Dublin often do not preside over a retrial, although on occasion the parties will agree between themselves that the same judge should sit and that rulings given in the earlier trial will be abided by. It is also agreed that, in other venues, judges will generally recuse themselves from the retrial upon an application being made. It is not, however, suggested that these practices are the result of any perception that there is a binding legal principle. This judgment deals with the question whether there is, or should be, any such principle, and does not question the merits of the general practice.
Background
5. For the purposes of this appeal it is not necessary to consider the merits of the case made by either the prosecution or the defence in the trial and a very brief summary of the issues will suffice.
6. One of the offences with which the appellant was charged was that of holding or disposing of waste in a manner which caused or was likely to cause environmental pollution (under the Waste Management Act 1996 as amended), while the other related to the keeping of prohibited goods used or intended to be used for removing prescribed markers from mineral oils (contrary to the Finance Act 1999). Given the nature of the charges, some of the significant evidence for the prosecution was given by officials from the local authority and from authorised Revenue officers, as well as officers of the Garda Síochána.
7. The case for the prosecution rested largely upon evidence gleaned from searches, carried out in certain commercial premises that were leased in the name of a company with which the appellant was associated. The initial entry into the premises was by members of the Garda Síochána, who were searching for stolen property and were in possession of a warrant issued by a District Judge. One issue in the voir dire was the basis upon which that warrant was obtained. It was claimed on behalf of the appellant that the garda who obtained the warrant deliberately misled the District Court judge about the source and extent of the information in her possession relating to the allegedly stolen property. She was also challenged in relation to certain remarks that she alleged were made by the appellant during the search. Her credibility and the reliability of her evidence were thus put firmly in issue.
8. In the course of their search the gardaí found items and material which gave rise to a suspicion that diesel laundering had been carried out in the premises. Contact was made with local authority officials and Revenue officers. Part of the defence case was that the garda search warrant was invalid, and it was argued that entry by such officials at the invitation of the gardaí would have been unlawful. However, their evidence was that they had, as a matter of law, their own statutory powers of entry into premises (such premises not being a dwelling) under, respectively, waste management and customs legislation. The challenge to their evidence in the voir dire therefore focussed heavily on the factual question whether those powers had actually been exercised, or whether the officials had simply entered on foot of a garda invitation. Again, the credibility of these witnesses was clearly put in issue.
9. Arising from certain of the evidence given in the voir dire, counsel for the appellant applied to have the case withdrawn from the jury, arguing that it would be an affront to justice to permit the trial to continue.
10. The trial judge rejected the challenge to the warrant, holding that there was nothing in the evidence to impugn the honesty or good faith of the garda who had obtained it, or her truthfulness to the District Judge. He described the allegation that the District Judge had been deliberately misled as “quite astounding”, and as being based on vague suggestions of mala fides that he found to be absurd. He also rejected the challenge to the entry onto the premises by the officials, holding on the evidence that while they attended at the premises because of the invitation from the gardaí, they acted in exercise of their statutory powers. It was not the invitation that conferred the power.
11. It is clear from the transcript that the trial judge disapproved of some aspects of the defence cross-examination, and in the course of the ruling the word “disingenuous” was used in respect of one particular aspect which the judge considered had given rise to a risk of unfairness.
12. However, the judge did rule that the evidence of one authorised Revenue officer should be excluded. It had been discovered in cross-examination that this witness had signed a number of different versions of her statement, and she appears to have been unable to give an adequate explanation of what she had done or of the resulting discrepancies. The judge held that the integrity and probative value of her evidence had been significantly diminished, although he considered that the problem was due to ham-fistedness and confusion on her part rather than (as contended for by the defence) deliberate deceit or gross negligence. The problem, as he saw it, did not concern the gathering of the evidence, but the inability of the witness to put her information into the form of a statement. However, he felt the point had been reached where he had to express the court’s disquiet by excluding her evidence from the trial.
13. The trial judge did not accept the proposition of defence counsel that the evidence of this witness contaminated the entire prosecution case, or that it presented an “enormous crisis of truth”, or that it brought the administration of justice into disrepute or would otherwise justify a direction to acquit having regard to the principles discussed in DPP v. J.C. [2017] 1 IR 417. He considered that it would be wrong to prohibit the trial, or to direct an acquittal, on the basis of the incompetence of one witness.
14. The jury having failed to agree on a verdict, the defence subsequently applied to the trial judge (in November 2019) to recuse himself from the retrial. It was argued that since he would be the trier of fact in the voir dire that would be required in the retrial, there was a risk that he might be unconsciously influenced by evidence, impressions of witnesses or his prior determinations. In those circumstances, it was submitted, a reasonable observer could reasonably conclude that if he made the same decisions as he had in the first trial, it would be due to such unconscious influence or to predetermination. No suggestion was made of subjective bias, or of unfairness in the conduct of the trial. Counsel did not, in making his application, expressly refer to the terms used by the judge in the course of his ruling.
15. The judge refused the application, referring to his declaration as a judge, his commitment to the presumption of innocence, and his duty to act impartially and in accordance with fair procedures. He also referred to his ability to disregard evidence previously heard and to the fact that the jury would be the ultimate decision makers and might take a different view of the evidence to his own.
The High Court
16. The appellant’s application for judicial review relief was in part grounded on an affidavit sworn by his solicitor, who repeated the argument that, in the light of the legal rulings and findings of fact made in the first trial, a reasonable objective observer would reasonably conclude that the trial judge would be “influenced by or unconsciously act in accordance with” his previous decisions, and would therefore have a reasonable apprehension that the issues in the voir dire would not receive a fair hearing by reason of pre-judgment. In those circumstances the appellant would be deprived of the fundamental constitutional right to have his trial heard by an impartial judge presiding over an impartial jury.
17. It was accepted on the appellant’s behalf that there was no general rule that a judge who had heard a particular case, or issue in a case, ought to be precluded from rehearing the matter. However, it was argued that a different situation arose where the judge had made findings as to credibility. As an alternative, it was submitted that any decision on recusal should take into account the nature and extent of the judge’s involvement in the determination of issues, the nature of those issues, and any observations or words of the judge that might be relevant.
18. The respondent submitted that this case differed from most of those dealt with in the authorities on bias, in that the trial judge in a criminal case was not the ultimate trier of fact. The voir dire, which involved a question of law, was held in the absence of the jury in order to avoid prejudicial material being heard by them, but judges were more capable than juries of putting potentially prejudicial evidence out of their minds. Where the evidence given in a voir dire was subsequently adduced in the trial, the jury was entitled to reach a different view on it to that formed by the trial judge.
19. MacGrath J. noted that the question whether a case should go to the jury must be determined by a trial judge as a matter of law. The determinations made by a judge in relation to evidence were different to those made by a jury in that, while the decision to be made by the judge was whether the evidence was admissible and capable of belief, the jury’s decision was whether or not to actually believe it. However, he did not consider that this division of responsibilities between the judge and jury necessarily led to the conclusion that the principles relating to bias were any less relevant since, in determining the issues of law in this case, the judge had been required to assess the evidence of the witnesses and make findings on their honesty.
20. MacGrath J. considered that it must always be a question of degree whether the circumstances were such that a trial judge should, as a matter of law, recuse himself or herself from a further hearing. He observed that in this case counsel, when applying to the trial judge to recuse himself, had not specifically drawn attention to the phraseology used by him in his ruling. While counsel was not to be criticised for this, it would have been preferable had he done so. However, the expressions used by the judge were but a factor in the case, and were not, in the circumstances, to be seen as central.
21. The primary concern, as MacGrath J. saw it, was the perception of pre-judgment in a case where the same or similar arguments would be raised before the same judge by the same counsel in a voir dire in a retrial. An objective and informed reasonable bystander might very well conclude that the same outcome was likely, and that the ruling of the judge would be influenced by what had gone before. However, “influence” was not the same as pre-judgment. The test was posed in the following terms at para. 40:-
“What this court has to consider is whether [the bystander] might reasonably apprehend that despite the undoubted ability, experience and expertise of the judge’s mind to compartmentalise and apply itself in a fresh manner on a fresh day to the same or similar issue concerning the credibility and admissibility of the same evidence or witness, that there is a risk that the matter may be, has been, or is likely to be, prejudged?”
22. MacGrath J. considered that the views of experienced practitioners, such as the appellant’s solicitor, could not be dismissed lightly when considering this question. However, in his view the significant factors were that the jury was the ultimate arbiter of guilt or innocence, that the legality of the rulings made in the previous trial was not in issue in the judicial review proceedings, and that there was no estoppel in respect of an issue determined in a voir dire. Not without some hesitation, he found that in those circumstances, while it might be wise for a trial judge to step aside in view of a previous involvement, the failure to do so did not amount to an error of law.
The Court of Appeal
23. After the High Court judgment and pending the hearing of the appeal in the Court of Appeal, counsel renewed the recusal application to the Circuit Court judge, in part for the purpose of remedying the omission noted by the High Court and drawing his attention to the language used by him in the course of the ruling. The Court of Appeal did not find it necessary to seek production of the transcript of this application, since it post-dated the High Court judgment. However, the judgment summarises what transpired, and the parties are agreed that the summary is correct. While still refusing to recuse himself, the trial judge was at pains to point out that he was not being and had never intended to be in any way critical of the manner in which the defence was conducted. The Court of Appeal confirmed that from its own view of the papers there was nothing that gave rise to the slightest concern that there was anything untoward. On the contrary, the defence was a “doughty” one and had been advanced with considerable ability and determination.
24. The President noted at the commencement of his judgment that the question raised by the case was not only interesting and significant, but novel. The grounds of appeal were distilled down to one question - whether, in circumstances such as those in this case, the informed and reasonable bystander would conclude that an accused would not receive a fair trial owing to the failure of the trial judge to recuse himself or herself.
25. In answering this question, the President observed that the authorities were not of particular assistance given that the issue had not previously arisen for consideration. However, he cited the following passage from the decision of the Privy Council in Stubbs v. The Queen [2019] AC 868 (“Stubbs v. The Queen”):-
“…the fair-minded and informed observer does not assume that because a judge has taken an adverse view of a previous application or applications, he or she will have pre-judged, or will not deal fairly with, all future applications by the same litigant. However, different considerations apply when the occasions for further rulings do not arise in the same proceedings, but in a separate appeal.”
26. In the instant case, the Court of Appeal considered that the context of the application for recusal was relevant. The defence had made an application in the trial to withdraw the case from the jury. That entailed discharging a very heavy burden. Any reasonable and informed observer would have formed the impression that the trial judge engaged with the issues and dealt with them in a particularly careful and conscientious manner. Any reasonable and informed observer would have no doubt that the judge would approach the task of presiding at a retrial, including conducting a voir dire if called upon to do so, with complete fairness.
27. However, the Court was conscious that the question was not whether the judge would strive to be fair, but whether his involvement in the earlier trial meant that he could not bring an open mind to bear. Counsel on behalf of the appellant had limited his argument to cases where the credibility of witnesses was in issue, but the fact that a judge had made findings on credibility could not in itself give rise to a presumption that he was precluded from sitting on a retrial. It was pointed out that the credibility in question was not that of the appellant, but that of prosecution witnesses. In the Court’s view, the significance of the distinction lay in the fact that a party whose credibility had been rejected in an earlier trial, perhaps in strong terms, might be concerned that the same judge would not be able to put that out of his mind and would be likely to come to the same conclusion again. That consideration did not apply with equal force where the complaint made by a party was that an attack on the credibility of the other side’s witnesses had not succeeded.
28. In conclusion, it was held that the objective bystander would not have any reasonable basis to have concerns in this case about the fairness or impartiality of any retrial presided over by the same judge. It was true that the bystander might feel that he could predict the outcome with some confidence, but that was often the case, and did not mean that the judge was not bringing a fair and open mind to bear.
Issues in the appeal
29. In the course of case management the parties agreed that the following two questions fell to be addressed in the appeal:-
a) In the context of a re-hearing of a criminal trial, what are the guiding principles that must be borne in mind when a recusal application is made before a judge who has already made findings relating to the credibility, reliability, honesty, integrity and motivations of witnesses?
b) In the context of the appellant’s case, applying those guiding principles, was recusal necessary to ensure that justice was seen to be done?
Submissions
30. It may be noted that the parties are, for the most part, agreed on the general principles relating to bias and pre-judgment, and have mostly referred to the same authorities to illustrate their arguments. As the agreed statement of the issues demonstrates, the question really comes down to the position of a trial judge who has made rulings in a voir dire based on contested oral evidence.
31. In addressing the Court of Appeal judgment, the appellant points to certain observations that he contends were made in error. He emphasises, firstly, that the case has not been made on the basis of any argument that the trial judge had not been careful or conscientious, or had not sought to balance fair trial rights with proper considerations as to the administration of justice.
32. It is also submitted that the Court of Appeal erred in according significance to the fact that some evidence was excluded by the trial judge - the evidence in question was, it is said, in itself of little or no significance in the case, and the reasonably informed observer must be taken to be aware of that. The defence argument being made in the trial went beyond the content of that evidence, and was concerned with the witness’s explanation for the discrepancies and the impact of that explanation on other parts of the prosecution case. It is further submitted that the Court of Appeal erred in attaching significance to the fact that the credibility findings related to prosecution witnesses and not to the appellant, in circumstances where the voir dire was all about the credibility, conduct and motivation of the prosecution witnesses.
33. The appellant submits that the judgment of the Court of Appeal does not deal with what he considers to be certain crucial issues that arise in respect of some of the evidence given in the voir dire, and the rationale of the trial judge for his findings in relation to that evidence. For example, he maintains that he had succeeded in demonstrating that the garda evidence about the search warrant supported the defence position that the District Judge had been misled. The use of the word “disingenuous” had also been shown to be incorrect, but the Court of Appeal had simply noted that this word was “infelicitous”. Similarly, the appellant contends that the evidence did not support the trial judge’s findings in respect of the entry into the premises by the Revenue officers, or his view that the witness whose evidence was excluded had merely been confused or incompetent in putting her statement together.
34. It is argued that it is both possible and desirable for the Court to consider these matters, on the basis that in order to apply the test for recusal it is necessary for this Court to appreciate the nature and circumstances of the findings of fact and the nature of the assessment of evidence leading the Judge to make those findings. The appellant contends that this proposition is exemplified by the fact that the Court of Appeal held that the exclusion of evidence and the use of critical language about the witness’s state of “confusion” showed that a reasonable observer would have no grounds for concern. The appellant maintains that this finding is untenable in the light of the evidence that emerged in the voir dire.
35. The appellant has offered a detailed analysis of the evidence to support these propositions. However, he does not ask this Court to determine whether the Circuit Court trial judge was correct in his findings of fact. He emphasises that he does not seek relief on the basis that the judge’s findings were irrational but argues that the determinations arrived at required an assessment of credibility and other nuanced factors. It is in that context, having regard to the nature of the issues of fact and the language in which the conclusions were expressed, that it is said that the apprehension of unconscious pre-judgment arises to the extent that the test for objective bias is met.
36. The appellant draws a distinction in principle between a decision of a trial judge in relation to a document, such as a warrant, and a decision based on contested oral evidence. The distinction is said to lie in the fact that the document will not change, and the same legal arguments will be available on appeal. By contrast, the words spoken by a witness, or the demeanour of that witness, may change in the circumstances of a retrial. However, the assessment of a witness is a subjective process, resulting in a finding that can rarely be successfully appealed. It is submitted that if a judge determines an issue on the basis of contested oral evidence, there will be grounds for concern that the same judge in the retrial will be affected by his or her recollection and perception of the first trial. It is submitted that, in consequence, the case for recusal is particularly strong where there is contested oral evidence.
37. It is also suggested that in some circumstances the test for recusal could be met where the decision of the trial judge does not involve an assessment of contested oral evidence. For example, a judge’s comments might give grounds for concern that that judge would not have an open mind on a legal issue or a new authority. However, it is not contended that the fact that it may well be anticipated that the same judge will make the same decision on exactly the same evidence in the retrial, as was made in the first trial, is in itself a ground for judicial review of a refusal to recuse. The predictability of a particular judge’s legal determinations is in an entirely different category from circumstances where the judge’s decision on contested oral evidence can be predicted. This is said to be particularly so when the decision can be predicted because the judge ruled in emphatic or vehement terms - on the same issues, hearing the same witnesses, in the same case– but in a different trial.
38. On the question as to the appropriate guiding principles, the appellant proposes the following as being based on well-established jurisprudence:
a) The first duty of a judge is to hear and determine cases in accordance with law in circumstances in which justice is seen to be done.
b) Judges are expected to do their duty to hear cases.
c) That duty, and considerations of administrative convenience, must yield to the constitutional obligation to ensure that justice is not only done but is seen to be done.
d) The test is objective: would a reasonable, well-informed observer reasonably form a view that the judge had predetermined an issue of importance in the retrial?
e) The mere fact that a judge has made decisions in regard to a criminal trial on indictment is not of itself sufficient to require recusal in the retrial. However, in any such case, unless the parties raise no objection, it may be preferable and, depending on the circumstances, necessary for a different judge to preside over the retrial.
f) Where the judge in a trial on indictment has made an assessment and determination on a matter such as the credibility, reliability, honesty, integrity, and motivations of witnesses on an issue of potential significance in the retrial, a recusal application should be granted; it is not necessary to show that particularly robust language was used by a judge in making such assessments and determinations, but such language may make the need for recusal more obvious.
g) Impugned comments, decisions and conduct by a trial judge must be considered in totality and in context.
39. It is stressed by the appellant that these propositions are put forward as guiding principles rather than rules, and that everything must ultimately depend on the facts of a case. In particular it is accepted that the first part of proposition (f) probably cannot, as a matter of practicality, be applied in an automatic fashion. However, the appellant sees it as an appropriate default position, in circumstances where, as a matter of practice, trial judges very often do recuse themselves from retrials.
40. In principle, the respondent agrees with propositions (a) to (d), and with (g), but takes issue with what is termed the “broadness and conditionality”, of (e) and (f), and the lack of legal authority thereof. In particular, the respondent adopts the view expressed by Lord Bingham in Locabail (UK) Ltd v. Bayfield Properties Ltd [2000] 2 WLR 870 to the effect that it is “dangerous and futile” to enumerate a list of criteria to be applied to a recusal application. What may be initially described as “guiding principles” risk becoming rules if expressed in mandatory terms, and could lead to a situation where, for example, a judge who hears an interim application in a family law case, or an interlocutory application in plenary proceedings, could not deal with the substantive hearing of the case. The position taken in relation to the agreed questions is that the same principles apply as apply to recusal applications generally, and that there are no specific rules in relation to judges who have made findings or rulings in a voir dire.
41. The respondent maintains the argument made in the High Court to the effect that the trial judge and jury have different roles, and the jury is the arbiter of the ultimate facts in issue. As far as rulings in a voir dire are concerned, the principle that issue estoppel has no role in criminal proceedings is pointed to as precluding any view that the result of a second voir dire on the same issue will be a foregone conclusion. Apart from any other consideration, different evidence may be called, or the evidence may emerge in a different way. If the appellant does not succeed in persuading the trial judge to accept his view of the evidence, he can appeal any error of law if convicted. He should not be entitled to what is characterised as “pre-emptive relief”. The contention that an appeal is the appropriate remedy is seen as strengthened by the fact that the appellant has, in his written submissions, directed considerable criticism to the merits of the rulings made in the voir dire.
42. It is contended that the appellant’s argument is based on an assumption that the trial judge is required to adjudicate on the honesty and reliability of prosecution witnesses generally. That is the role of the jury. It was however necessary for the judge to embark on an assessment of the credibility of the witnesses in this case, because the approach of the appellant in the voir dire would, if successful, have involved a finding by the judge that the garda deliberately misled a District Court judge. The trial judge was not satisfied that there was a basis for such a finding, and he was entitled to express his view on the issue. It remained open to the jury to reach a different conclusion.
43. The respondent takes issue with the argument of the appellant that no significance should be attached to the exclusion of some of the prosecution evidence. It is submitted that the appellant’s view of the importance of that evidence is not relevant - the point is seen as lying in the fact that the reasonable observer would have seen the judge strongly criticising prosecution evidence. In any event, the respondent objects to consideration of the merits of the rulings of the trial judge as part of the appellant’s case. It is argued that the appropriate forum to determine such issues is an appeal if the appellant is convicted. For the purposes of this appeal, the conduct of the preceding trial by the trial judge is only relevant insofar as it affects the recusal application. The findings of fact by the trial judge are themselves not the proper subject matter of these judicial review proceedings.
44. Insofar as the potential distinction between rulings based on oral evidence and rulings based on documents may be of relevance, the respondent points out that the conduct of a voir dire by a trial judge might involve the assessment of oral evidence, documentary evidence or a combination of both. Given the static nature of documentary evidence, it might be that a voir dire primarily concerned with documents is likely to lead to a similar result when conducted on several occasions by different judges. It is submitted that the predictability of the outcome of a voir dire is not a useful test as it potentially includes an assessment of the strength of the point litigated and the previous decisions of the judge in other cases, neither of which are relevant in the consideration of bias.
Discussion
45. For the purposes of this discussion it may be necessary to emphasise that the concept of objective bias does not imply any criticism of the decision-maker whose decision is the subject of challenge. Subjective, or actual, bias arises where the decision is said to have been actually motivated by personal reasons such as animosity, or self-interest rather than being based on a fair evaluation of facts and the application of the law to those facts.
46. Objective bias, on the other hand, is more concerned with perception, and hence the central question is whether in given circumstances the neutral, reasonable observer would apprehend that the decision-maker might be biased. It carries no implication of personal wrongdoing or impropriety. As Keane C.J. pointed out in Orange Communications Limited v. The Director of Telecommunications Regulation (No. 2) [2000] 4 IR 159 (“Orange”), the distinction is encapsulated in the famous dictum that justice should not only be done but should manifestly and undoubtedly be seen to be done (Lord Hewart L.C.J. in R. v. Sussex Justices; Ex p. McCarthy [1924] 1 KB 256). Subjective bias is concerned with the first part of this principle, objective bias with the second. Although the word “bias” can carry very negative connotations in the ears of judges, it is necessary that they should not, therefore, be overly sensitive when applications for recusal are made on grounds of objective bias. It is not to be equated with an allegation of bad faith.
47. In a nutshell, the appellant’s case is that the reasonable observer would apprehend that he will not get a fair retrial because the trial judge has previously assessed certain prosecution witnesses as being honest and credible, and, because of the human process by which such an assessment is reached, is likely to be affected by his previous decisions and unlikely to come to any other conclusion in a retrial. As Birmingham P. observed in the Court of Appeal, this particular issue is novel and not directly covered by any of the authorities.
48. It is necessary, firstly, to consider the appellant’s argument that the Court should, in determining whether or not the test for recusal has been met in this case, take into account the particular findings and rulings made on particular evidence in the trial. My own view is that it should not. One obvious, practical reason is that this matter is going to be retried, and it is no part of the function of this Court, in the context of the single issue to be determined in this particular appeal, to express or appear to express views as to how issues in the forthcoming trial should or should not be decided. To suggest even tentative agreement or disagreement with a particular analysis of a particular ruling or finding could have the effect of prejudicing the proper conduct of the forthcoming trial.
49. Apart from this consideration, there is an issue of principle. The rulings made by the trial judge in the trial were either legally correct or they were not. In an appeal against conviction, an appellate court would consider whether the rulings were erroneous in the light of the evidence and the applicable legal rules. In a judicial review, a court could be asked to quash rulings for irrationality or other vitiating error. However, this Court is not acting in either of those capacities. The sole issue before us is whether the trial judge was wrong to refuse to recuse himself, in a case where the argument is not grounded on the nature of the rulings made by him, but on a question of principle.
50. The suggestion that the Court should, in reaching its conclusions on this question, consider the allegedly infirm rulings in the light of the evidence may seem innocuous, but runs the risk of breaching one of the principles identified in the judgments in Orange. The appellant in the instant case is not, of course, asking the Court to analyse the alleged errors for the purpose of inferring bias, which did occur to some extent in Orange. In very brief summary, the High Court judge in that case accepted that objective bias could, without any other evidence, be inferred on foot of her finding that the decision under challenge contained so many errors, and was so unreasonable, that there was no other explanation or justification.
51. In this Court, Keane C.J. pointed out that in none of the authorities relating to bias did the courts concern themselves in the slightest degree with the merits of the decision under consideration. Further, in cases about objective bias, the courts did not concern themselves in the slightest degree with the manner in which the judge or tribunal reached the decision. If there was a reasonable apprehension of bias the decision must be set aside for that reason alone, but the authorities lent no support whatsoever to the proposition that a court could infer, from the establishment of a number of errors in the impugned decision or in the process leading up to it, that the decision was vitiated by the existence of bias that could be equated with objective bias.
52. For the purposes of this appeal, it is relevant that Keane C.J. went on to say that it was not surprising that there was no authority supporting such a proposition, because it would be entirely contrary to principle. Errors of fact or law could be corrected on appeal, and unreasonable decisions, or decisions made after a process that breached the requirements of fair procedures, could be set aside in judicial review. However, to say that errors and unreasonableness led to a further presumption that the decision was totally vitiated by bias was “a non sequitur of truly startling dimensions”.
53. The other judgments in Orange deal with this issue in similar terms. Barron J.’s view was that there had to be evidence establishing the factor - a relationship, interest or attitude - that actually influenced, or might reasonably be perceived as influencing, the decision in question. Murphy J. agreed - the existence of the extraneous impermissible factor must be proved on the balance of probabilities, (although if that was done its effect would be presumed). An erroneous decision, “even one made in the context of criticism of counsel for the unsuccessful party”, was not evidence of bias, and nor was evidence of a series of adverse findings in a particular case. Murphy J. also endorsed the view of the United States Supreme Court (in National Labour Relations Board v. Pittsburgh S.S. Co. (1949) 337 U.S. 656) that a “total rejection” of an opposed view could not of itself impugn the integrity or competence of a trier of fact”. Geoghegan J. observed that there was no authority for the proposition that bias could be established by proving cumulative error all one way, or from the nature of the decisions made - there had to be proof of circumstances outside the actual decisions made.
54. In simple summary, the argument made in the instant appeal that consideration should be given to the alleged errors in the trial judge’s findings, is that the judge erred in his assessment of the evidence in the first trial and, human nature being what it is, is likely to come to the same conclusions, and thus make the same errors, in the future. The question, then, is whether it is possible for a party to establish bias in the sense of pre-judgment in respect of a future hearing, by the demonstration that errors occurred in the past, notwithstanding that such errors could not ground a finding of past bias.
55. I consider that this question must be answered in the negative. When a court is asked to quash a decision that has already been made, on grounds of bias, it is not concerned with the merits of the decision. It seems to me that it would be illogical, therefore, to examine those merits for the purpose of determining whether the decision-maker will be biased in the future. To do so would be to equate error with bias, and to ask the superior courts to prevent the commission of future errors on the basis of a claim of bias. I do not propose, therefore, to go into any further detail in relation to the events of the first trial.
56. The form of bias said to exist in this context is pre-judgment, or pre-determination. The appellant does not suggest that the trial judge pre-judged the issues in the voir dire in the original trial. The decision of the Court therefore, in reality, depends upon its answer to the question whether this is a case where it can be said that the forthcoming trial, if presided over by the same judge, will be or will appear to be tainted by pre-judgment because he has previously made decisions regarding the credibility of witnesses giving contested oral evidence. The question could be put in the terms used by Charleton J. in Kelly v. Minister for Agriculture [2021] IESC 23 (“Kelly”) - could there in those circumstances be a reasonable apprehension that the judge will have “decided an outcome in advance”, such that the voir dire will become “a tendentious exercise in self-confirmation”?
57. In Orange, Barron J. considered that it was inherent in the nature of bias that the extraneous factor giving rise to an apprehension of bias should have existed before the hearing. It should be noted, however, that in Kelly O’Donnell J. cautioned against the over-reading of this aspect of that case. The important point, in his view, was that the factor giving rise to a claim of bias must be extraneous to the correctness of the impugned decision. There are indeed some cases where the element of pre-judgment was shown to have existed before the impugned decision was made - for example, in Orange, Murphy J. cited Berger v. United States (1921) 255 U.S. 22, where (in a case where the defendant was a German American) the trial judge said that one would have to have “a very judicial mind, indeed” not to be biased against German Americans, and that he would prefer the company of a safe-blower to that of the defendant. Barron J. referred to re Watson, Ex p. Armstrong (1976) 50 A.L.J.R. 778, where a judge dealing with an interlocutory application stated that, having read the affidavits, he would not be prepared to accept the uncorroborated evidence of either party in the substantive hearing.
58. However, as Clarke J. pointed out in P.(A) v. His Honour Judge McDonagh [2009] IEHC 316, there also is a form of pre-judgment that arises, not from any pre-existing factor, but in circumstances where the decision maker indicates that he or she has reached a conclusion on a question in controversy at a time prior to it being proper to reach such a conclusion. That will be the case where the decision maker uses language that might cause the reasonable observer to believe that he or she has excluded any realistic possibility of coming to any other conclusion, at a stage when further evidence or argument is to be expected. The observer will consider that a premature rush to judgment has occurred, although there may be no particular animus towards the party who stands to lose.
59. Since a finding of objective bias does not require a finding that a judge has actually pre-determined an issue, the test, as set out in O’Neill v. Beaumont Hospital Board [1990] I.L.R.M. 419, is whether a reasonable observer would apprehend that the appellant’s chance of a fair and independent hearing of the trial does not exist, by reason of pre-judgment of the issues involved in the anticipated voir dire. In Dublin Well Woman Centre v. Ireland [1995] 1 I.L.R.M. 408. Denham J. (giving the unanimous decision of the Court) stated that this test was fundamentally the same, although differently phrased, as that formulated by Lord Denning MR in Metropolitan Properties Co. (FGC) Ltd v. Lannon [1969] 1 QB 577:-
“…In considering whether there was a real likelihood of bias, the court does not look at the mind of the justice himself or at the mind of the chairman of the tribunal, or whoever it may be, who sits in a judicial capacity. It does not look to see if there was real likelihood that he would, or did, in fact favour one side at the expense of the other. The court looks at the impression which would be given to other people. Even if he was as impartial as could be, nevertheless if right-minded persons would think that, in the circumstances, there was a real likelihood of bias on his part, then he should not sit. And if he does sit, his decision cannot stand…Nevertheless there must appear to be a real likelihood of bias. Surmise or conjecture is not enough… There must be circumstances from which a reasonable man would think it likely or probable that the justice, or chairman, as the case may be, would, or did, favour one side unfairly at the expense of the other. The court will not inquire whether he did, in fact, favour one side unfairly. Suffice it that reasonable people might think he did. The reason is plain enough. Justice must be rooted in confidence; and confidence is destroyed when right-minded people go away thinking: The judge was biased.”
60. It will be noted that, in either formulation of the concept, the focus is on whether the reasonable observer would apprehend unfairness by reason of pre-judgment - the concern is not whether the result in relation to a particular issue is predictable. In this regard, as Fennelly J. pointed out in O’Callaghan v. Mahon [2008] 2 IR 514 (“O’Callaghan v Mahon”), the apprehensions of the actual affected party are not relevant. That party is, obviously, not to be equated with a fair-minded and neutral observer.
61. The analysis in the judgment of the Privy Council in Stubbs v. The Queen is helpful in this regard. The sequence of events there was that Isaacs J. had presided over a murder trial in the Bahamas, in the course of which he made a number of legal rulings. The trial had to be aborted before reaching a jury verdict, and the accused were retried before a different judge. Again, a number of rulings were required to be made in the trial. Having been convicted, the accused appealed, and a question arose as to whether it was proper for Isaacs J. to sit in the Court of Appeal. That Court considered the objection to be unsustainable. As summarised in the judgment of the Board of the Privy Council, the reasons were that the aborted trial over which Isaacs J. had presided had taken place some seven years earlier and the passage of time should assuage any apprehension of bias; that since the trial had been aborted there was no record of what rulings he had made - “the slate had been wiped clean” and the rulings were not in issue in the present appeal; that Isaacs J. would not be sitting alone but as one of a panel of three; that it could not realistically be said that his involvement in the earlier trial precluded him from approaching the appeal with an open mind; and that the courts should be astute to guard against “judge-shopping”, which could be a blight on the proper administration of justice.
62. Giving the judgment of the Board, Lord Lloyd-Jones JSC referred to the principle that bias as a result of pre-determination or pre-judgment would be a ground for recusal, citing dicta by Sedley LJ in Amjad v. Steadman-Byrne (Practice Note) [2007] 1 WLR 2484 - “The appearance of bias includes a clear indication of a prematurely closed mind” - and Longmore LJ in Otkritie International Investment Management Ltd v Urumov [2015] CP Rep 6 - “The concept of bias…extends further to any real possibility that a judge would approach a case with a closed mind or, indeed, with anything other than an objective view; a real possibility, in other words, that he might in some way have ‘pre-judged’ the case.”
63. At paragraph 16 of the judgment, Lord Lloyd-Jones referred to the position of a judge who has to make successive rulings in the course of the same litigation:-
“A judicial ruling necessarily involves preferring the submissions of one party over another. However, it is obviously not the case that any prior involvement by a judge in the course of litigation will require him to recuse himself from a further judicial role in respect of the same dispute. In the great majority of such cases there will simply be no basis on which it could be suggested that the judge should recuse himself, notwithstanding earlier rulings in favour of one party or another, and there will often be great advantages to the parties and to the administration of justice in securing judicial continuity. The issue will only arise at all in circumstances where prior involvement is such as might suggest to a fair-minded and informed observer that the judge’s mind is closed in some respect relevant to the decision which must now be made. It is not possible to provide a comprehensive list of factors which may be relevant to this issue which will necessarily depend on the particular circumstances of each case.”
64. This passage clearly covers the position of a judge who has made interim or interlocutory rulings in the course of litigation - in the absence of evidence of bias, there is no reason why the same judge should not hear the substantive matter and indeed it may be desirable to have continuity. The reasons for distinguishing between the position of a trial judge tasked with a retrial and the position of Isaacs J., who intended to sit as an appellate judge, is discussed in paragraph 31. The prosecution had made an argument to the effect that there could have been no objection if Isaacs J had sat as the trial judge in the retrial:-
“This is correct. Indeed, it is often the case that, following an aborted trial, a retrial takes place before the same judge and a different jury. However, this does not assist the respondent. While this process may involve the same judge revisiting the rulings he made at the first trial, this is essentially a repetition of one stage of the judicial process in circumstances where the earlier rulings were rendered of no effect by the aborted trial. There is no prejudice to a defendant in these circumstances. If a previous ruling against the defendant is repeated at the retrial the defendant is in no worse a position and, if there are good grounds, he will be able to appeal the ruling to an independent and impartial appellate tribunal. On the other hand, where, as here, one is concerned with an appeal, very different considerations apply. An appellant is entitled to be heard by an independent and impartial appeal tribunal without any appearance of bias by way of pre-determination or pre-judgment. For the same reason, the common case of a judge who has to make successive rulings in the same proceedings (see para 16 above) is not analogous to the present case. In the former the high desirability of judicial continuity is an important factor, whereas in the present case this consideration is entirely absent.”
65. I agree with this analysis, and with the view that a trial judge should not be compelled to recuse himself or herself from a retrial simply on the basis that it can be anticipated that the same issues will require to be ruled on, or the same witnesses heard, as in a previous trial that did not come to finality. It is true that the assessment of witnesses giving contested evidence in a voir dire may to some extent be a subjective process, but it is a process carried out within legally defined parameters, by a legally-trained and experienced lawyer, and is one that requires a reasoned decision. As I said earlier, decisions made by trial judges are either correct or incorrect in terms of the applicable law, and the criminal justice system has established procedures to remedy errors.
66. Having said that, the practice, which appears to be widespread, of not sitting in retrials unless it is administratively necessary, or the parties request it, may often be the prudent course for trial judges. The concept of judicial continuity may, depending on the circumstances, have some value in this context, and presumably that is why the parties will sometimes agree to having the same judge preside on the basis that they will abide by the rulings from the earlier trial. However, the desirability of continuity is not as great as in the case of a sequence of interim, interlocutory and substantive hearings. Furthermore, it can be difficult to participate in any capacity in a retrial and to hear the same witnesses being cross-examined on the same issues (or indeed to conduct such cross-examinations). While I do not accept the argument that judges are bound by any general principle to recuse themselves simply because they had to rule on witness evidence in a voir dire, I would agree with MacGrath J. that it is sometimes wise to accede to a recusal application.
67. It remains possible that a trial judge may, in making the necessary decisions in a trial, give rise to an apprehension of bias through an unfortunate choice of language. As Fennelly J. said in O’Callaghan v. Mahon, objective bias may be established by showing that the decision maker has made statements which, if applied to the case at issue, would effectively decide it or which show prejudice, hostility or dislike towards one party or his witnesses. The use of robust language should not, however, necessarily give rise to such apprehension - since the fact-finders in a criminal trial must have regard to the standard of proof beyond reasonable doubt, a finding in favour of the prosecution after a voir dire will necessarily be expressed with confidence and in terms that exclude reasonable doubt.
68. Unnecessarily trenchant or insulting criticisms should however be avoided. It is certainly the case that judges should avoid using opprobrious words about a party, or his or her professional representatives, unless the situation appears to call for it. For example, the word “disingenuous” is best not used unless the speaker is aware of, and intends to convey, the full connotations of that word when applied to a professional advocate. Carelessness with language may lead to unintended affront and upset, and might in some circumstances lead to a justified application for recusal. However, some latitude has to be allowed for the vicissitudes of a live trial. And, as with judges, counsel should not be overly sensitive. Situations like this may often be resolved by asking the judge for clarification. Thus, for example, it was made clear by the trial judge in this case that he had not intended to criticise the manner in which the defence was run, when counsel returned to him after delivery of the High Court judgment. I would not consider (and counsel does not contend) that the manner in which the rulings in the trial were delivered would, without more, meet the standard of bias.
69. In the circumstances, I do not consider that the appellant has established that the principles relating to bias require any particular rule in the case of retrials, whether or not the original trial involved the determination of any issue by the trial judge. I would therefore dismiss the appeal.
Result: Dismiss