Groves, M --- "The Rule Against Bias" [2009] UMonashLRS 10

Last Updated: 6 May 2011

The Rule against Bias

Matthew Groves[1]

The rule against bias is one of the two pillars of natural justice. The hearing rule governs the procedural features of decision making. The bias rule governs the attitude or state of mind of the decision maker. This article examines the foundation of the bias rule and the fiction of the “fair minded and informed observer” by whose conclusions claims of bias are determined. The article considers whether the objective test that supposedly underpins the notion of the fair minded and informed observer does little more than provide a cloak for the subjective views of the judges who apply that doctrine.

Introduction

The rule against bias is one of the twin pillars of natural justice. The first pillar --the hearing rule --requires that people whose rights, interests and expectations may be affected by a decision should be given sufficient prior notice and an adequate chance to be heard before any decision is made. The bias rule is the second pillar of natural justice and requires that a decision-maker must approach a matter with an open mind that is free of prejudgment and prejudice. Although the bias rule originated in the courts, and was for many centuries applied only to courts and judges, it has now become a rule of almost universal application. The rule against bias applies to a vast range of decision-makers including tribunals,[2] statutory [486] authorities,[3] court officials,[4] juries,[5] government ministers,[6] local councils,[7] prison officials,[8] bureaucrats[9] and more senior government officials,[10] coronial inquiries,[11] and even private arbitrators.[12]

As the bias rule has expanded to include a great range of decision-makers it has also become more flexible. The courts have repeatedly stressed that the bias rule must take account of the particular features of the decision-maker and wider environment to which the rule is applied. The Supreme Court of Canada has explained that “the contextual nature of the duty of impartiality” enables it to “vary in order to reflect the context of a decision maker's activities and the nature of its functions.”[13] There are many similar judicial pronouncements which stress that the bias rule is context sensitive.[14] At the same time, however, the courts have adopted a single test to determine applications for bias --that of the fair minded and informed observer.[15] This fictitious person provides a vessel in which the courts can impart as little or as much knowledge as is required to provide context. In many cases the courts imbue the fair minded and informed observer with remarkably detailed knowledge and considerable understanding and acceptance of decision-making. This approach begs the question of [487] whether the fair minded and informed person is a neutral observer or little more than the court in disguise.

This article examines how the courts apply the rule against bias and the knowledge that they impute to the fair minded and informed observer. It will be argued that the apparently objective nature of the fair minded and informed observer is often a mirage and that judges frequently impose their own subjective opinions rather than those of any objective person. One consequence of such criticisms, which is considered in the final section of this article, is whether an objective test for claims of bias is inherently flawed because it will inevitably be so strongly influenced by the judges who apply it that the test can never hope to acquire the level of objectivity to which it lays claim. The article also analyses recent developments in the law of England and Australia of potential relevance to the Hong Kong Special Administrative Region (HKSAR). The most notable is the decision by Australian courts to overturn the longstanding rule of automatic disqualification for pecuniary interest in favour of a single test for all claims of bias and advocates a similar change in the HKSAR. But first it is useful to rehearse the foundations of the bias rule and the two forms of bias identified by the courts.

What are the Foundations of the Bias Rule?

The precise origins of the rule against bias are unclear but there is longstanding common law authority to support the principle that the decisions of courts could be set aside, or that judges might not be permitted to preside, if the judge was thought not to be impartial.[16] Most of these early cases involved judges who had a direct interest in proceedings before them. By the early seventeenth century such a conflict of interest was regarded by English courts as “against right and justice and against natural equity.”[17] The more recent historical analysis of these cases has focussed on the evolution of the principle of automatic disqualification for pecuniary interest. More particularly, this analysis has focused on whether a rule of automatic disqualification is as well settled within the common law as its supporter [488] claim.[18] This recent emphasis on the origins of automatic disqualification is not necessarily helpful to a wider understanding of the rule against bias because any rule of automatic disqualification is ultimately no more than a specific application of the wider rule against bias.

The principle upon which the bias rule has been founded in modern times can be traced to Lord Hewart's famous statement that “justice should not only be done, but be seen to be done.”[19] On this view, appearances are important. Justice should not only be fair, it should appear to be fair. Lord Hewart's statement signalled the rise of the modern concern with the possible apprehension that courts and judges might not appear to be entirely impartial, rather than the narrower problem that they might in fact not be impartial. The importance of the appearance of impartiality has become increasingly linked to public confidence in the courts and the other forms of decision-making to which the bias rule applies.[20] This rationale of the bias rule also aligns with the objective test by which it is now governed because the mythical fair minded and informed observer, whose opinion governs the bias rule, is clearly a member of the general public.[21] Accordingly, the views attributed to the general public provide both the justification for and content of the bias rule.

The requirement of impartiality which lies at the heart of the bias rule is often confused with the related issue of independence but there is a clear difference between the two. In Ebner v Official Trustee[22] the High Court of Australia explained that “[B]ias, whether actual or apparent, connotes the absence of impartiality.” But the Court also noted that bias “may not be an adequate term to all cases of the absence of independence.”[23] Baroness Hale [489] of the House of Lords also drew attention to the distinct and related nature of impartiality and independence in Gillies v Secretary of State for Works and Pensions[24] when she explained:

“Impartiality is not the same as independence, although the two are closely linked. Impartiality is the tribunal's approach to deciding the cases before it. Independence is the structural or institutional framework which secures this impartiality, not only in the minds of the tribunal members but also the perception of the public.”[25]

The underlying point of this reasoning is that impartiality is a concept generally directed to specific instances of decision-making, while independence is an institutional concept that governs the wider structures within which decision-makers act. This personal/institutional distinction provides a convenient taxonomy to categorise all of the significant recent English decisions on bias. Some cases have focussed on the institutional arrangements governing decision-makers, in which case a claim of bias is determined by examining whether the constitution and operation of the relevant body is sufficiently distanced from the executive that it may provide a fair hearing for human rights purposes.[26] When such claims succeed, they have led to considerable institutional reform. In other cases the claim of bias has been directed at a particular decision in a more personal sense, by an argument that the decision-maker has a personal interest or connection that may affect impartiality.[27] When claims of this nature succeed, the remedy is not [490] far reaching and is usually limited to ensuring that the challenged decision-maker is removed from the case at hand.

Professor Lucy has suggested that impartiality and independence may not be so easily separated because impartiality is normally embedded in both a decision-maker and the process by which that person functions.[28] On this view, impartiality cannot simply be an attitude in the part of the decision-maker. It must also extend to the wider process within which that person functions.[29] Some judges have suggested that judicial independence and impartiality are sufficiently interwoven that both should be regarded as constitutional requirements.[30] In my view, there are several reasons why the rule against bias, or any similar principle of impartiality in the part of decision-makers, should not be rested on constitutional foundations. Some of these reasons are of general application while others have particular relevance for Hong Kong.

First, if the bias rule was transformed into a constitutional principle applicable to courts and judges, the basis upon which it would extend to nonjudicial decision-makers such as tribunals and administrative officials would be unsettled. Public confidence in the legal process, to which the bias rule is commonly anchored, could be diminished if the rule was limited to some decisions affecting legal rights but not others. Public confidence would also be diminished if the basis or content of the bias rule differed radically between courts and other forms of decision-making.[31] It is important to note that rigid distinctions between the constitutional requirements applicable to courts and tribunals in Hong Kong law should not necessarily be drawn. There is some authority suggesting that a “court” for the purposes of Article [491] 35 of the Basic Law includes a tribunal that exercises judicial power.[32] According to this view, fundamental or constitutional requirements applicable to courts might also extend to some tribunals.[33] In this author's view, this possibility does not provide a strong reason to anchor any requirement of impartiality to a constitutional foundation because the extension of some requirements applicable to courts to some tribunals does not remove the basic problem that could arise by the application of those requirements to some but not all non-judicial decision makers.

Secondly, if the bias rule was constitutional in nature, its precise features might depend very much on the form and structure of constitutions within individual jurisdictions.[34] Those principles might even vary within a single jurisdiction.[35] Any such fragmentation of doctrinal principles would be at odds with the increasingly transnational nature of the law governing the bias rule, which would be particularly undesirable in the HKSAR in light of the strong emphasis of the courts on comparative analysis in the development of public law in the post-1997 environment.[36]

Thirdly, any suggestion that the rule against bias should be founded on constitutional principles provides little guidance for jurisdictions that lack a written or those which exist within such a unique constitutional structure [492] as the HKSAR.[37] Although Chief Justice Li has stated that the Basic Law is similar to other constitutions because “it distributes and delimits powers ...”.[38] it should not be assumed that the Basic Law enables an acceptance within the HKSAR of the separation and distribution of power adopted in the constitutional arrangements of other jurisdictions. The Basic Law clearly bears Westminster influences but it is equally clear that some core Westminster concepts such as the separation of powers have no place in the PRC Constitution and occupy a somewhat uncertain position within the law of the HKSAR.[39] Sir Anthony Mason has suggested that principles associated with the separation of powers doctrine developed in other jurisdictions must be approached with particular caution in Hong Kong. Sir Anthony explained the reason for this caution in the following terms:

“The Basic Law incorporates a separation of powers. So far, however, the courts of the HKSAR have not had occasion to consider what the doctrine may entail in Hong Kong. It would not follow that the Basic Law, when construed in light of its context and the preservation of the English common law by Article 8 of the Basic Law, necessarily mandates a separation of powers that conforms to either the United States or Australian model.”[40]

This conclusion is consistent with the reasoning of Peter Wesley-Smith, who has argued that Hong Kong's constitutional evolution is increasingly accommodating a particular conception of the separation of powers, though the precise features of this doctrine are not yet entirely clear.[41]

[493] These various arguments indicate that the separation of judicial power and its attendant requirements are conceptually different from any constitutionally based requirement of impartiality. There might also be good reason why the two ought to remain separated, at least to some extent. This conclusion does not mean that the principles governing the rule against bias cannot or should not evolve so as to reflect the particular legal framework of Hong Kong. Bokhary PJ drew attention to this possibility in Financial Secretary v Wong[42] when his Honour noted that judicial review was “a rapidly developing area of the law.” Statements such as this make clear that the law of judicial review in Hong Kong is, like the body of doctrine arising from the Basic Law, in a state of flux.

Actual v Apprehended Bias

Bias may take many different forms but the main distinction is between actual and apprehended bias. A claim of actual bias requires proof that the decision-maker approached the issues with a closed mind or had prejudged the matter and, for reasons of either partiality in favour of a party or some form of prejudice affecting the decision, could not be swayed by the evidence in the case at hand.[43] A claim of apprehended bias requires a finding that a fair minded and reasonably well informed observer might conclude that the decision-maker did not approach the issue with an open mind.[44]

These differences between actual and apprehended bias have several important consequences. Each form of bias is assessed from a different perspective. Actual bias is assessed by reference to conclusions that may be reasonably drawn from evidence about the actual views and behaviour of the decision-maker. Apprehended bias is assessed objectively, by reference to conclusions that may be reasonably drawn about what an observer might conclude about the possible views and behaviour of the decision-maker.[45] [494] Each form of bias also requires differing standards of evidence.[46] A claim of actual bias requires clear and direct evidence that the decision-maker was in fact biased. Actual bias will not be made out by suspicions, possibilities or other such equivocal evidence. In the absence of an admission of guilt from the decision-maker, or, more likely, a clear and public statement of bias, this requirement is difficult to satisfy.[47] A claim of apprehended bias requires considerably less evidence. A court need only be satisfied that a fair minded and informed observer might conclude there was a real possibility that the decision-maker was not impartial.[48]

An allegation of actual bias places a court in an invidious position because it requires a serious adverse finding of a personal or subjective nature to be made against the decision-maker. The courts are naturally reluctant to make such findings, which may explain the high evidentiary standard applicable to claims of actual bias. A court that upholds a claim of apprehended bias is not required to make an adverse finding against the decision-maker. It can instead make the more palatable finding that a reasonable observer, though not necessarily the court, might conclude that the decision-maker was not impartial and go no further.[49] Although the courts frequently stress that a claim of apprehended bias will not be upheld lightly,[50] it is clear that [495] the evidence required to sustain such a claim is not nearly so strict as it is for actual bias. In Locabail (UK) Ltd v Bayfield Properties Ltd[51] the Court of Appeal of England explained this issue in the following terms:

“The proof of actual bias is very difficult, because the law does not countenance the questioning of a judge about extraneous influences affecting his mind; and the policy of the common law is to protect litigants who can discharge the lesser burden of showing a real danger of bias without requiring them to show that such bias actually exists.”[52]

If a successful claim of either form of bias is sufficient to set aside a decision there is little incentive for a party to incur the more onerous and potentially sensitive requirements involved in a claim of actual bias. Justice Kirby acknowledged this practical issue when he conceded that a party who sought to raise a claim of actual rather than apprehended bias would be “foolish ... to assume a heavier obligation when proof of bias from the perceptions of reasonable observers would suffice to obtain relief.”[53] This admission highlights the overlap between the two forms of bias. A case that might support a claim of actual bias may often be argued upon the ground of apprehended bias simply because the latter is easier to establish. Callinan J of the High Court of Australia acknowledged this point in Johnson v Johnson[54] when he conceded that “in some exceptional cases a submission of apprehended bias may be no more than a polite fiction for no doubt unintended, unconscious and ultimately unprovable, but nonetheless actual bias ...”.[55]

The Rise of the Fair Minded and Informed Observer

The fair minded and informed observer is a relatively recent judicial device to determine claims of bias. For a long time questions of bias were determined subjectively by the courts. This approach was affirmed most recently in 1993 by the House of Lords in Lord Goff explained it was “... unnecessary, in formulating the appropriate test, to require that the court should look at the matter through the eyes of the reasonable man, because the [496] court ... personifies the reasonable man.”[56] This test for bias is a subjective one that depends squarely upon the opinion of the court. The views of the parties or the general public have no role in determining a claim of bias. The Lords also made clear that the standard was one of a “real danger” as opposed to a “real likelihood” or “real suspicion”.

This subjective, judge focussed, approach was pointedly rejected a year later by the High Court of Australia in Webb v R.[57] The High Court adopted an objective test, which determined a claim of bias by reference to the “reasonable apprehension on the part of a fair-minded and informed observer ...”.[58] A key advantage of this approach is that it aligns the content of the rule against bias with the underlying rationale of the rule because an objective test gauged through the eyes of a member of the public enables claims of bias to be determined by reference to the very public whose confidence the bias rule ultimately seeks to secure. This key benefit of the objective test for bias is apt to be undermined if the reasonable observer is attributed with too much specialist knowledge. In the Webb case Mason CJ and McHugh J made clear that the knowledge that could be imputed to this observer must necessarily be limited if the person was to remain a hypothetical member of the public rather than the court masquerading as such. They explained:

“... the public perception of the judiciary is not advanced by attributing to a fair-minded member of the public a knowledge of the law and the judicial process which ordinary experience suggest is not the case.”[59]

But they added:

“That does not mean that the trial judge's opinions and findings are irrelevant. The fair-minded and informed observer would place great weight on the judge's view of the facts. Indeed, in many cases the fair minded observer would be bound to evaluate the incident in terms of the judge's findings.[60]

According to this view, the objective observer ought to be informed but not too much so. In Locobail (UK) Ltd v Bayfield Properties Ltd[61] the English Court of Appeal suggested that the Australian and English tests might usually be similar:

[497] “Provided that the court, personifying the reasonable man, takes an approach which is based on broad common sense, without inappropriate special knowledge, the minutiae of court procedure or other matters outside the ken of the ordinary, reasonably well informed member of the public, there should be no risk that the courts will not ensure both that justice is done and that it is perceived to be done.”[62]

A notable feature of this passage is not simply the apparent belief that judges embody reasonableness but also that they are able to distance themselves from the detailed specialist knowledge about the legal system which lies at the core of their judicial duties. The following year it became clear that a differently constituted Court of Appeal doubted the ability of judges to step aside from detailed knowledge of the legal system, though they offered a more mundane reason for the adoption of an objective test for bias based upon the views of the public rather than judges.

The English Court of Appeal reviewed European law Re Medicaments and Related Classes of Goods (No 2)[63] and concluded that a test of a “reasonable apprehension of bias” would be more consistent with the requirements of European law. The Court concluded that European law supported a “modest adjustment” of the English test for bias. This adjusted test required a court to:

“... first ascertain all the circumstances which have a bearing on the suggestion that the judge was biased ... then ask whether those circumstances would lead a fair-minded and informed observer to conclude that there was a real possibility, or a real danger, the two being the same, that the tribunal was biased.” [64]

When the House of Lords subsequently accepted this adjustment in Porter v Magill[65] it also affirmed that the fair minded observer would take account of the circumstances of the case at hand. It did not, however, provide any significant guidance on exactly how much the fair minded observer should be taken to know.

The precise standing in the HKSAR of the test adopted in Porter arguably remains unsettled. When the Court of Final Appeal considered the issue in Deacons v White & Case Ltd Liability Partnership[66] it pointedly declined to make an authoritative ruling on whether the bias rule should be determined by reference to the view of the court (the Gough test) or the [498] fair minded an informed observer (the test from Webb and Porter v Magill). The CFA concluded that it was “unnecessary ... to comment definitively on the applicable test” because all parties to the appeal agreed “that the test in Hong Kong is the reasonable apprehension of bias test ...”.[67] The weight of authority suggests that Porter is widely applied within the lower courts of Hong Kong. In some cases the Porter test has been applied without significant discussion by the court.[68] In others, the court has applied the test espoused in Porter, though without specific reference to that case.[69] In my view, these cases indicate that there is a clear weight of authority in Hong Kong law suggesting that the fair minded observer espoused in Porter v Magill is now widely accepted and would, in an appropriate case, almost certainly be adopted by the CFA. The next section of this article indicates that this test opens a new difficulty in the bias rule.
Stretching the Limits of the Well Informed and Reasonable Observer

In the Webb case the High Court of Australia cautioned that the gap between actual and apprehended bias narrowed as the knowledge that was credited to the fair minded and informed observer increased. Despite this caution, many cases indicate that the courts attribute unusually detailed knowledge to the fair minded observer. In particular, courts often impute quite detailed knowledge about legal procedures and technicalities to the fair minded observer. The Court of Appeal of Hong Kong, for example, has suggested that the fair minded observer would know and understand when a judge was “determining an interlocutory matter, not on the basis of exercising a discretion but on the basis of whether evidence of documents are relevant to issues in the proceedings ...”.[70] The Supreme Court of Canada has imputed the observer with a “full knowledge of the Quebec municipal court system, including all of its safeguards.”[71] The Court of Appeal of England has held that the observer would understand the difference between a [499] draft judgment which was issued subject to possible alteration, as opposed to one that was final and binding.[72] The Court also held that the observer would also know and understand the provisional nature of the decision as explained by previous judicial decisions.

Such cases beg the question of precisely who the fair minded observer is supposed to be? Is it a member of the public imputed with a level of knowledge, as many cases suggest, or it the judge who has imputed that knowledge to the fair minded observer? Kirby J was a longstanding critic of this feature of cases on bias. In Johnson v Johnson[73] he reasoned that the hypothetical person was a “reasonable member of the public” who was “neither complacent nor unduly sensitive or suspicious.”[74] Although this statement has been widely cited in Hong Kong and other jurisdictions,[75] the wider context of Kirby J's reasoning generally has not been.[76] His Honour questioned the growing tendency of courts to assign many specific qualities and remarkably detailed knowledge to the fair minded observer. In a passage worth quoting in full, Kirby J explained:

“The attributes of the fictitious bystander to whom courts defer have therefore been variously stated. Such a person is not a lawyer. Yet neither is he or she a person wholly uninformed and uninstructed about the law in general or the issue to be decided. Being reasonable and fair minded, the bystander, before making a decision important to the parties and the community, would ordinarily be taken to have sought to be informed on at least the most basic considerations relevant to arriving at a conclusion founded on a fair understanding of all the relevant circumstances. The bystander would be taken to know common place things, such as the fact that adjudicators sometimes say, or do, things that they might later wish they had not, without necessarily disqualifying [500] themselves from continuing to exercise their powers, the bystander must also now be take to have, at least in a very general way, some knowledge of the fact that an adjudicator may properly adopt reasonable efforts to confine proceedings within appropriate time limits and to ensure that time is not wasted. The fictitious bystander will also be aware of the strong professional pressures on adjudicators (reinforced by the facilities of appeal and review) to uphold traditions of integrity and impartiality. Acting reasonably, the fictitious bystander would not reach a hasty conclusion based on the appearance evoked by an isolated episode of temper or remarks to the parties of their representatives, which was taken out of context.” [77]

His Honour warned that courts should hesitate to attribute too much knowledge to the fair minded observer because “for a court simply to impute all that was eventually known to the court to an imaginary person ... would only be to hold up a mirror against to itself.”[78] Kirby J subsequently expanded upon this criticism in Smits v Roach[79] where he complained that the “fictitious postulate” of the reasonable, intelligent, fair-minded lay observer had “been stretched virtually to snapping point”. His concern was not simply the ever growing list of qualities attributed to the hypothetical person but the failure of judges who engage in such reasoning to recognise that they sought to cloak a very personal and subjective approach to claims of bias with the guise of the reasonable and objective observer.[80] He suggested that the attribution of so many features to the hypothetical observer was:

“... provided by the judges to remind themselves, the parties and the community reading their reasons that the standard that is applied is not simply the reaction of the judges, at trial or on appeal, to a particular complaint. It is, as far as it can be, an objective standard: one aimed at emphasising the undesirability of idiosyncratic and personal assessments of such matters. As the cases show, in such decisions different judges can reflect different assessment and reach different conclusions. The fact that this is so should make contemporary judges aware that, ultimately, they themselves hide behind a fiction and pretend that it provides an entirely objective standard by which to measure the individual case.”[81]

[501] There is clear force in these observations. The detailed and ever expanding knowledge ascribed to the hypothetical person clearly has a subjective element. It reflects the qualities that the judges believe that reasonable and objective people ought to have rather than what such people really do have. This interaction between an approach that is ostensibly objective yet clearly infused with the subjective views of the judges might seem an unpalatable blend, but what are the alternatives? It may be argued that any test for bias will, if the rule makes recourse to the importance of public confidence in the administration of the law, inevitably comprise a mixture of the objective and subjective issues. The objective element of the test naturally arises from the recourse of any such test to notions of public confidence and the views of the public that such an approach necessarily involves. But such an approach will inevitably incorporate a subjective element because the judges who administer the test will inevitably draw upon their own views.

The criticisms made by Kirby J have particular force in respect of the knowledge that the courts impute to the fair minded observer about the culture and traditions of the legal profession. This issue usually arises in cases where a claim of bias is founded upon professional connections between a judge and the lawyers acting for one party. These problems can arise in various forms. A judge may have shared chambers with a barrister who now appears before that judge. The judge may, prior to judicial appointment, have been regularly briefed by a solicitor who now acts for a party appearing before that judge. The judge may while at the Bar have engaged a barrister as a junior many times, and even served as a mentor to that barrister, and subsequently have that barrister appear before him or her. The common point in these and other possible examples is that a judge can possess a prior association with, or knowledge of, a lawyer by reason of the judge's earlier professional life. Such issues do not generally appear to trouble the courts, even though the connection between the judge and one or more lawyers may have been close and longstanding.

Many years ago the Court of Appeal of New South Wales reasoned that the fair minded observer would not be troubled by longstanding or close connections between judges and lawyers because a significant level of public knowledge and acceptance of such past connections within the legal system was “built into the system.”[82] This approach which is reflected in many decisions across many jurisdictions enables courts to conclude that the fair minded observer would understand and accept many practices within the legal profession that arguably might surprise the general public. Perhaps the [502] best recent illustration is the decision of the English Court of Appeal in Taylor v Lawrence.[83] In that case the court accepted that the fair minded observer would not entertain a reasonable apprehension of bias against a judge who had engaged the solicitor acting for a party in a case before him to draft his will (a fact he disclosed to the other, unrepresented, party). This finding was not affected by the fact that the same lawyers provided free assistance to enable the judge and his wife to amend their wills the night before the judge delivered judgment in the instant case (a fact the judge did not disclose to the other party). The Court of Appeal held that these facts would not support a reasonable apprehension of bias in the fair minded observer because that person would accept that the behaviour complained of had occurred against a background of “legal traditions and culture” that served to disseminate and maintain ethical standards and, in turn, worked to “promote an atmosphere which is totally inimical to the existence of bias”.[84]

Although the facts of Taylor v Lawrence may be unusual, the willingness of the courts to accept that legal traditions and culture can provide a secure basis upon which to defeat claims of bias is not. In a recent Hong Kong case a judge presiding over a dispute that had arisen during arbitration proceedings considered Taylor v Lawrence and concluded that the reasoning of the Court of Appeal of England “applies equally to the legal traditions and culture of Hong Kong ...” She also added that “the same statement can be made of the wider dispute resolution circle.”[85] This reasoning invites several comments. First, it suggests that the belief by the courts that fair minded observer would understand the peculiar traditions of the legal profession is a widely held one. One logical consequence of the widespread acceptance of this view that the fair minded observer would surely be taken to understand the particular traditions of the legal profession of the jurisdiction in which that person is invoked. In other words, the fair minded observer in Hong Kong would understand the legal culture and traditions of Hong Kong. The same person located in Singapore or New Zealand would have a similar local knowledge and so on. It should also be noted that the suggestion quoted above (that the reasoning in Taylor v Lawrence applies to dispute resolution circles) makes clear that the fair minded observer will be taken to have a knowledge and understanding of technical issue and traditions extending beyond traditional adversarial litigation in the courts. That person will apparently have a similar knowledge of issues related to arbitration, but what other areas might this possibility extend to? Mediation and other forms of [503] alternative dispute resolution? But what other areas might this reasoning extend to?

A final comment that can be made about the tendency of courts to impute knowledge about legal culture and traditions to the fair minded observer is that the courts have given no real guidance on exactly how much will be imputed to the fair minded observer. Will that person be imputed with knowledge of every aspect of the culture and traditions of the legal profession and system? Or will that knowledge be selected? The lack of judicial guidance on this issue highlights the difficulty in ascertaining precisely what the fair minded will be taken to know which, in turn, draws attention to a key problem that flows from the willingness of the courts to impute detailed specialist knowledge about the legal profession to the fair minded observer. This tendency enables the courts to imbue the fair minded observer with the peculiar cultural values of lawyers and judges. This undercuts the apparently objective nature of the bias test by transforming the person by whose judgement the bias test is gauged from a fair minded and informed member of the public to a fair minded and informed member of the legal profession. From that point, it is a small step for the fictional person to become the judge who is considering the claim of bias.

Another significant problem with the tendency of the courts to ascribe detailed knowledge and acceptance of legal culture to the fair minded observer is that it becomes harder for courts and judges to recognise the institutional problems that may be part of that culture. Baroness Hale of the House of Lords acknowledged this problem when she suggested that, if the fair minded observer was an “insider” to the legal system he or she “would run the risk of having the insider's blindness to the faults that outsiders can so easily see.”[86] An English commentator addressed this problem in more detail shortly after the decision in Taylor v Lawrence was delivered when he criticised the willingness of courts to impute significant knowledge to the fair minded observer in the following terms:

“... not only had detailed knowledge of the administration of justice been imputed to the observer, but the cases ... have also come dangerously close to equating that knowledge with approval of the relevant practice. This approach is inherently conservative, and is not conducive towards taking a ‘fresh look’ at the administration of justice”[87]

[504] In my view, the subtle distinction between knowledge and approval of a practice or tradition, particularly of the legal system, may not be capable of easy resolution. The difficult nature of this issue should not be used as a reason for unthinking criticism of judges. It is easy for outside observers to identify and criticise the internal practices of the legal system, and to do likewise for other areas of decision-making such as public sector decision-making within government, but the criticisms of external observers may also be criticised from the so-called inside view. Just as those who work within the legal system may be argued to be less able to examine the traditions and practices of that system from a fresh perspective, those who do not work within the legal system are surely less able to fully understand the traditions and practices of the legal system and the effect those matters may have. Neither perspective can be perfectly informed. The views reached by those who are within or outside the legal system will be informed by different experiences and perspectives. Striking an appropriate balance between those different perspectives is a difficult exercise. If one accepts the difficult nature of that task and that this difficulty may be assisted by an understanding of the issues and the wider institutional environment in which those issues occur, the particular experience of judges is a legitimate and useful influence in determining claims of bias.

The Anomaly of the Continued Existence of Automatic Disqualification

The principle of automatic disqualification is generally traced to Dimes v Grand Junction Canal,[88] although it was noted above that the point at which the automatic nature of disqualification for pecuniary interest became a settled principle of law is disputed.[89] For present purposes, four inter-related points may be made about the Dimes case. First, automatic disqualification was applied without question in many cases after Dimes.[90] A separate but logically related point is that the application of the rule of automatic disqualification for pecuniary interest appeared to harden over time. The shareholding of the Lord Chancellor in Dimes was worth a fortune but later cases made clear that a pecuniary interest would almost always lead to disqualification whatever its value. The important quality was the nature of the interest rather than its size. Thirdly, automatic disqualification provided a simple, perhaps blunt, solution to claims of bias by reason of pecuniary [505] interest but offered no real solution to claims founded on other interests. Finally, any principle of automatic disqualification provides an exception to the test of the fair minded and informed observer because the automatically nature of disqualification essentially imposes a presumptive finding of bias without reference to the fair minded observer.

The House of Lords did not share these concerns when it decisively affirmed automatic disqualification in R v Bow Street Magistrate Ex parte Pinochet (No 2)[91] (Pinochet No 2). Put briefly, the problem in that case was that Lord Hoffman who had determined an appeal for the extradition of the former Chilean dictator (Pinochet) was closely associated with Amnesty International (Amnesty). Amnesty had mounted a long campaign seeking to hold dictators legally accountable for their actions. Pinochet had been a prime target of this campaign. Amnesty was granted leave to intervene in the extradition proceedings and argued strongly in favour of extradition. Lord Hoffmann was not a member of Amnesty but was instead an unpaid director of a charity that Amnesty wholly controlled. His Lordship's connection to Amnesty was close and, in light of Amnesty's intervention, he was virtually an unpaid director of an intervening party. The case could have been disposed of by any formulation of the bias test, which would have lead to the disqualification of Lord Hoffmann on the ground that his close association to a party to the case gave rise to a reasonable apprehension of bias. The House of Lords instead held that Lord Hoffmann was subject to automatic disqualification. In effect the Lords did not simply affirm the rule of automatic disqualification but extended the reach of that doctrine to the perceived interest of Lord Hoffmann and, more generally, non-pecuniary interests. Lord Browne-Wilkinson stated the rationale of this extension in the following terms:

“If the absolute impartiality of the judiciary is to be maintained, there must be a rule which automatically disqualifies a judge who is involved, whether personally or as a director of a company, in promoting the same causes in the same organisation as is a party to the suit.”[92]

The House of Lords noted that Lord Hoffmann, Amnesty and the charity were separate legal entities. Lord Hoffmann was not, therefore, a party to the proceeding in a formal sense or a judge in his own cause but the Lords thought that the connection between Lord Hoffmann and Amnesty [506] could still be characterised as an interest that should lead to disqualification. This reasoning had the obvious advantage of allowing the House of Lords to avoid findings either way on actual or even ostensible bias. It had the added advantage of not having to argue that pecuniary involvement is more obviously disabling than unpaid involvement,[93] or the related issues of whether and how the effect of pecuniary and non-pecuniary interests might be compared against each other. But the extension of the rule of automatic disqualification opened up another can of worms. The House of Lords accepted that mere non-directorial membership of Amnesty could have required Lord Hoffmann's automatic disqualification.[94] This aspect of the decision poses problems for judges who are linked to public “causes”, even those with a lower profile or less vigorous commitment than Amnesty, if such links may trigger disqualification. This aspect of the Pinochet case highlights the main disadvantage of the extension of automatic disqualification, namely that the precise reach of this species of automatic disqualification remains unsettled.

Many English litigants quickly sought to test the limits of Pinochet. The result was a series of challenges to judges on many grounds, in the guise that the supposed interest of the judge was one that required automatic disqualification. In many cases the parties had investigated the habits and history of the judge in their search for supposedly incriminating evidence. The Court of Appeal sought to stamp out this undesirable practice through a dogmatic judgment in the Locobail case,[95] in which it outlined the various qualities that would never, or almost never, support a claim of bias. The Court held that bias could virtually “never” be founded upon a judge's gender, age, race, religion, class, wealth or sexual preference. A further list of factors that would “hardly ever” support a claim of bias included a judge's membership of professional, sporting or charitable associations, extra-judicial writings, the judge's political or social or educational background, any connections related to the judge's former chambers and, lastly, any Masonic associations.[96] The final list, which the court held would ordinarily support a claim of bias included family connections, personal friendships and dislikes, and any close professional relationships. This division of influences and interests into categories indicates the differing extent to which they may, according to the perception of a fair mined and informed observer, influence a [507] decision-maker. From this view, the categories devised by the Court of Appeal provide some insight into the potential reach of the rule of automatic disqualification after Pinochet case.

The High Court of Australia took a quite different view in Ebner v Official Trustee[97] when it considered two joined appeals concerning judges who held relatively small shareholdings in a bank that was a party to proceedings over which the judge presided. The High Court could have decided each case on the basis of a de minimis exception to automatic disqualification.[98] A majority of the court instead held that there was no separate rule of automatic disqualification when a judge held a direct pecuniary interest in a party to a proceeding over which the judge presided.[99] The majority essentially reinterpreted Dimes by holding that that case did not, upon close reading, support automatic disqualification.[100] The majority also rejected the proposition that pecuniary and other interests could or should be treated differently, or that the latter should lead to automatic disqualification. Gleeson CJ, McHugh, Gummow and Hayne JJ stated:

“a rule of automatic disqualification would be anomalous. It is in some respects too wide, and in other respects too narrow. There is no reason in principle why it should be limited to interests that are pecuniary, or why, if it were so limited, it should be limited to pecuniary interests that are direct. This is illustrated by the problem that concerned the House of Lords in Pinochet (No 2). The concept of interest is vague and uncertain. It is not logical to have one rule for interest and a different rule applying to disqualification for association.”[101]

According to this approach, the question of whether a pecuniary or other form of interest may give rise to a reasonable apprehension of bias should be [508] determined by a single test. Gleeson CJ, McHugh, Gummow and Hayne JJ explained that test as follows:

“First, it requires the identification of what is said might lead a judge (or juror) to decide a case other than on its legal and factual merits. The second step is no less important. There must be an articulation of the logical connection between the matter and the feared deviation from the course of deciding the case on its merits. The bare assertion that a judge (or juror) has an ‘interest’ in litigation, or an interest in a party to it, will be of no assistance until the nature of the interest, and the asserted connection with the possibility of departure from impartial decision making, is articulated. Only then can the reasonableness of the asserted apprehension of bias be assessed.”[102]

Allegations of bias arising from a pecuniary interest must now be determined on a more reasoned basis than was the case with automatic disqualification. Although the court ultimately decides whether the connection between the interest and the apprehension of bias can be articulated to the relevant degree, the parties may now bear a heavier onus. A party cannot simply rest on a “bare assertion” of a pecuniary interest. A party who does not, or cannot, articulate the connection between the interest and the resulting apprehension, or least provide some basis to do so, risks the objection being dismissed as a “bare assertion”.[103]

Subsequent Australian cases suggest that the unitary test is not without difficulty. The requirement to articulate how the interest in question might lead a decision-maker to depart from acceptable standards of fairness is one over which courts can easily disagree.[104] Despite this problem, the unitary test adopted by the High Court of Australia has several advantages over a rule of automatic disqualification. One is that the adoption of a single test recognises that the key question in a claim of bias is not the form an interest might take but the effect it might have. A separate but related argument against automatic disqualification is that the extension of this rule in the Pinochet case implicitly accepts that some non-pecuniary interests, notably association to political or social causes, ought to give rise to automatic disqualification. At the same time, however, the Pinochet case provides no guidance as to when or why such associations will give rise to automatic disqualification. For this reason, automatic disqualification through association [509] with a political or social cause is little better than a ruler without a scale. It provides no clear guidance on why or when it should be used.

There is also a more practical reason in favour of the abolition of automatic disqualification for pecuniary interest. A separate category for pecuniary interest is difficult to defend in the complex world of global financial arrangements that has evolved since the Dimes case. Ownership of shares and other financial products was relatively uncommon when the Dimes case was decided but this is clearly no longer the case. The abolition of a rule of automatic disqualification enables judges and other decision-makers to hold shares in large publicly listed companies such as banks and other multinational companies. Such companies are frequently parties to many proceedings before the courts, as a consequence of their business. Shares in such companies, particularly very large ones, are also commonly held by individuals such as judges who have led a successful professional life. In my view, there is no reason in principle why judges or prospective judges should be forced to either divest themselves of a reasonably balanced share portfolio or refuse entirely to hear cases involving large corporate entities who are commonly litigants to proceedings in the courts.
The many pragmatic considerations in favour of a single test for claims of bias weighed heavily on the Court of Appeal of New Zealand in Muir v Commissioner of Inland Revenue.[105] The Court declined to settle the issue because it had not been fully argued by the parties but it accepted that “there are powerful arguments for simplicity and straightforwardness in this area of law, which has been somewhat devilled by contradictory approaches.”[106] The Court also noted that such a unitary test for bias would not weaken the traditionally strict approach to pecuniary interest because where “a judge has a direct pecuniary interest of anything more than the most minimal character, it is hard to see how the reasonable observer would not consider that to be bias.”[107] In my view, the same considerations provide a powerful basis for the courts of Hong Kong to reject a rule of automatic disqualification in favour of a single test applicable to all claims of bias along the lines that has occurred in Australia.

[510] Is an Objective Test for Bias Inevitably Flawed?

The Court of Appeal of England has defined bias as “a predisposition or prejudice against one party's case or evidence on an issue for reasons unconnected with the merits of the issue.”[108] This and similar expositions of the bias rule make clear that predisposition, attitudes and perhaps even some measure of prejudice are not necessarily impermissible. Scalia J of the Supreme Court of America acknowledged this possibility when he explained that prejudice consisted of a “favourable or unfavourable disposition that is somehow wrongful or inappropriate, either because it is undeserved, or because it rests upon knowledge that the subject ought not to possess.”[109] The implication in this passage is that predispositions may be permissible if they are warranted or somehow rational. Similarly, predisposition and other such qualities are not in themselves inappropriate. They only become so when they have no rationale basis or connection to the case at hand. The reason the law tolerates predisposition and other qualities that might appear to offend the rule against bias is practical. Professor Lucy argues that absolute total impartiality “might be possible but certainly not desirable.”[110] According to this view, judges who seek absolute impartiality:

“... would set aside what they know of human kind and their lives. The beings then judging us would know nothing at all of what standard human lives look or feel like or, knowing something, would completely ignore it. Expecting real judges to embody such an attitude would be to expect them to live debased lives. Just like us, their commitments and associated experiences make them the people they are; they serve to give judges both prior knowledge of human life and found various prejudgments and evaluations in their own lives.”[111]

There is considerable force in the argument that judges and other decision-makers cannot, and should not, be devoid of experience and the inevitable preconceptions that experience may bring, the point at which desirable experience becomes unacceptable baggage remains unsettled. The best known example remains the decision of the Supreme Court of Canada in RDS v [511] R.[112] In that case a black judge presiding in a juvenile court in Nova Scotia Court acquitted a black juvenile defendant of assaulting a white police officer. The only direct evidence of the incident was that of the police officer and the defendant. Each offered starkly different evidence, which the judge sitting alone could not reconcile. When the prosecutor closed his case he asked the rhetorical question of why the police officer would lie. The judge adverted to this point when she acquitted the defendant, by drawing upon her own knowledge of the well known racial tensions in the local area and police behaviour. She suggested that police officers sometimes overreacted while on duty and subsequently lied about such behaviour in court. Although the judge made no such finding against the officer who had testified before her she drew upon her general knowledge to conclude there was sufficient reason to find reasonable doubt which obliged her to acquit the defendant.

A bare majority of the Supreme Court of Canada upheld the decision of the trial judge but did so in terms that caused considerable controversy. The minority Justices held that the reasoning of trial judge displayed both prejudice and prejudgment because she had effectively accepted that all police officers were racist liars. This, they concluded, clearly breached the rule against bias. The majority Justices did not simply reject the suggestion that the trial judge was biased, they also held that the experiences and associated preconceptions she held were an entirely permissible influence. The majority Justices reasoned that all judges would inevitably possess a measure of experience and preconceptions based upon that experience, which they would naturally draw upon in their decision-making, and that this judge had essentially done so.

L'Heureux-Dube and McLachlin JJ, with whom Gonthier and La Forest agreed, accepted that “triers of fact will be properly influenced in their deliberations by their individual perspectives on the world in which the events in dispute in the courtroom took place.” They added that “[I]ndeed, judges must rely on their background knowledge in fulfilling their adjudicative function.”[113] They concluded that this experience and associated predispositions would not breach the rule against bias “so long as those experiences are relevant to the cases, are not based on inappropriate stereotypes, and do not prevent a fair and just determination of the cases based on the facts in evidence.”[114] Their Honours drew support from a publication of the Canadian Judicial Council, which explained the importance of judicial experience in the following terms: [512]

“... there is no human being who is not the product of every social experience, every process of education, and every human contact with those with whom we share the planet. Indeed, even if it were possible, a judge free of this heritage of past experiences would probably lack the very qualities of humanity required of a judge ... True impartiality does not require that the judge have no sympathies or opinions; it requires that the judge nevertheless be free to entertain and act upon different points of view with an open mind.”[115]

Although the RDS case concerned judicial decision-making, the principles espoused by the Supreme Court of Canada must surely be true for other decision-makers, such as tribunal members, ministers and bureaucrats. All decision-makers inevitably bring their own personal knowledge, and very often a good measure of institutional experience, to their task. It is difficult to accept that only judges can and should be informed by their wider experiences.

The approach favoured by the majority Justices in RDS has been described as “contextual judging”. At one level this term simply draws attention to the obvious connection between decision-making and the wider context within which it occurs. At a deeper lever, however, it implies that the personal values and experience of the decision-maker provide a legitimate part of that decisional context. The concept of contextual judging is arguably an inevitable consequence of the increasing importance placed on the need for a more representative judiciary. While this goal is clearly laudable, contextual judging has significant ramifications for the rule against bias because it confers legitimacy upon qualities in a decision-maker which might otherwise offend the bias rule. Experience and the preconceptions that it gives rise to may be permissible in the guise of contextual judging but there is always the possibility that preconceptions may be so strong that they constitute prejudgment.

The House of Lords recently took the point a step further in R v Abdroikov[116] where it considered several cases in which police officers had [513] served as jurors.[117] The proceeding before the Lords comprised several co-joined appeals against criminal convictions which raised the wider question of the extent to which the law could permit jurors to hold a range of differing views without infringing the rule against bias. Lord Mance noted that the representative nature of juries necessarily implied that there could be a divergence in the experience and views held by individual jurors within a single jury. He also suggested that any model of a reasonable and objective juror ought to accommodate “the widely differing characteristics, experience attitudes and beliefs” which could influence the deliberations of jurors, without those differing views “being cast as unreasonable.”[118] The difficult underlying issue which his Lordship did not directly confront is the point at which the views or experience of a juror diverge so far from the norm that they may be categorised as either unreasonable or sufficiently unrepresentative and, therefore, liable to breach the rule against bias.

Abdroikov is only one of a growing number of recent cases in which the courts have sought to balance differing experiences and views that is implicit in the representative nature of juries on the one hand, with the need to preserve the impartial nature of juries.[119] It is not yet clear precisely how much difference the courts might accept in the guise of the representative nature of juries. It is, however, clear that the courts believe that jurors and other decision-makers will take their duties seriously and may cast aside or control their possible prejudices and perform their duty fairly.

Concluding Observations

Some English commentators have suggested that Porter v Magill[120] represents a backward step because it prevents the courts from openly imposing their view as was possible under the Gough test.[121] The underlying point of these criticisms is that the Porter test simply requires courts to engage in a process similar to that advocated by Gough, though concealed weakly by an objective test. The logical consequence of these criticisms, whether it takes the form of a return to Gough or the adoption of a new formulation of the subjective tests that lay at the heart of Gough, is to allow the courts to [514] openly apply their own views in bias cases. The device of the fair minded and informed observer would be an inevitable casualty of this approach but, if one accepts the fictitious nature of this doctrine, that person exists in name only. Another possibility is to place greater weight on the views of the parties. The difficulty with this approach is that parties are invariably interested in the proceeding and could not, on any measure, provide a credible touchstone to determine claims of bias. Whatever problems may lie with undue reliance on the subjective views of judges must surely be doubly true of partisan parties. On this view, the hypothetical observer may be an imperfect device and may only be as fair or open minded as each judge allows, but this person remains the best vessel to test claims of bias.

---

[1] BA LLB (Hons), PhD; Senior Lecturer, Faculty of Law, Monash University, Australia. This article was written while the author was a Visiting Fellow at the Centre for Comparative and Public Law in the Faculty of Law at Hong Kong University. Thanks are due to Simon Young for his assistance and insight on the law of the HKSAR.

[2] Cheung v Insider Dealing Tribunal [2000] 1 HKLRD 807; Lawal v Northern Spirit Ltd [2003] UKHL 35; [2004] 1 All ER 187 (HL); Gillies v Secretary of State for Work and Pensions [2006] 1 All ER (HL); Grant v Teacher's Appeals Tribunal (Jamaica) [2006] UKPC 59. The rule also clearly extends to public sector disciplinary tribunals and their proceedings. See, eg, Rowse v Secretary for Civil Service [2008] HKCFI 549; [2008] 5 HKLRD 217 and Lam Siu Po v Commissioner of Police [2008] 2 HKLRD 27.

[3] PCCW-HRT Telephone Ltd v Telecommunications Authority [2008] 2 HKLRD 282 (telecommunications regulator); Re Duffy [2008] UKHL 4 (specialist body regulating street parades).

[4] R v Salford Assistant Committee Ex p Ogden [1937] 2 KB 1 (tribunal clerk).

[5] See, eg, R v Gough [1993] UKHL 1; [1993] AC 646 (HL); Webb v R [1994] HCA 30; (1993) 181 CLR 41 (HCA); R v Abdroikov [2007] 1 All ER 315 (HL).

[6] Minister for Immigration and Multicultural Affairs Ex p Jia (2001) 205 CLR 507 (HCA); Hot Holdings Pty Ltd v Creasy [2002] HCA 51; (2002) 210 CLR 438 (HCA); Imperial Oil Ltd v Quebec (Minister for Environment) (2003) 231 DLR (4th) 577 (SCC).

[7] Porter v Magill [2002] 2 AC 357; Man O'War Station Ltd v Auckland City Council (No 1) [2002] 3 NZLR 577 (UKPC); McGovern v Ku-Roing-Gai Council [2008] NSWCA 209.

[8] R (on the application of Al-Hasan) v Secretary of Statement for the Home Department [2005] UKHL 13; [2005] 1 All ER 927 (HL) (deputy prison governor exercising disciplinary powers).

[9] Minister for Immigration, Local Government and Ethnic Affairs v Mok (1994) 127 ALR 402 (ministerial delegate); Kwan v Victoria Legal Aid [2007] VSC 235 (bureaucrat).

[10] Bahadur v Secretary for Security [2000] 2 HKLRD 113 (accepting that the bias rule was applicable to administrative decisions made by the Secretary of Security of the HKSAR).

[11] R v Inner West London Coroner Ex p Dallagio [1994] 4 All ER 139.

[12] Pacific China Holdings Ltd v Grad Pacific Holdings Ltd [2007] 3 HKLRD 741; Jung Science Information Technology Co Ltd v ZTE Corporation [2008] 4 HKLRD 776.

[13] Imperial Oil Ltd v Quebec (Minister for Environment) (2003) 231 DLR (4th) 477.

[14] Minister for Immigration and Multicultural Affairs Ex p Jia (2001) 205 CLR 507 at 539, 551, 584 (distinguishing the standards expected of government ministers compared to other decision-makers); Bell v CETA (2003) 227 DLR (4th) 193 at 204-207 (distinguishing between the standards expected of courts and tribunals); PCCW-HKT Telephone Ltd v Telecommunications Authority [2007] HKCFI 129; [2007] 2 HKLRD 536 at 549 (distinguishing between an administrative authority and a tribunal); Allidem Mae G v Kwong Si Lin [2003] (HCLA 35/2002) at [39] (noting that the bias rule “must bear in mind the specific characteristics and actual circumstances of the Labour Tribunal”).

[15] An important distinction between some jurisdictions remains on automatic disqualification. A rule of automatic disqualification for pecuniary and other interests prevails in many jurisdictions but has been emphatically rejected in Australia.

[16] Earl of Derby's case [1572] EngR 116; (1613) 12 Co Rep 114; 77 ER 1390. See also Brookes v Rivers (Earl of) (1655-69) Hardes 503; 145 ER 569.

[17] Day v Savadge [1792] EngR 643; (1614) Hob 65; 80 ER 235 at 235. In that case an action in trespass was heard by the city officials against whom the claim was made. The decision was overturned by reason of the inherent conflict that the city officials faced.

[18] See, eg, Harry Woolf, Jeffrey Jowell and Andrew Le Sueur, De Smith's Judicial Review (London: Sweet and Maxwell, 6th edn, 2007) 501-502 where it is argued that the early bias cases are explicable on several grounds, including one of automatic disqualification for pecuniary interest, but that a firm rule of automatic disqualification was not established until the middle of the 19th century. A similar view is adopted in Abimbola Olowofoyeku, “The Nemo Judex Rule: The Case Against Automatic Disqualification” [2000] Public Law 456 at 456-458, though that author argues that a rule of automatic disqualification for pecuniary interest was not clearly established until early in the twentieth century.

[19] R v Sussex Justices Ex p McCarthy [1923] EWHC KB 1; [1924] 1 KB 256 at 259. In the same year, Aitkin LJ similarly remarked that “[N]ext to the tribunal being in fact impartial is the importance of its appearing so”: Shrager v Basil Dighton Ltd [1924] 1 KB 274 at 284.

[20] See, eg, Ebner v Official Trustee [2000] HCA 63; (2000) 205 CLR 337 at 363 (Gaudron J) (HCA); Re Medicaments and Related Classes of Goods (No 2) [2000] EWCA Civ 350; [2001] 1 WLR 700 at [83] (Eng CA); Lawal v Northern Spirit Ltd [2003] UKHL 35; [2004] 1 All ER 187 at [14], [21] (HL); Forge v Australian Securities Commission [2006] HCA 44; (2006) 229 ALR 223 at [66] (Gummow, Hayne and Crennan JJ) (HCA). See also Belilos v Switzerland [1988] ECHR 4; (1998) 10 EHRR 466 at [67] where the European Court of Human Rights explained that the bias rule, as it arose from Art 6 of the European Convention of Human Rights, was based upon the importance of “the confidence which must be inspired by the courts in a democratic society”.

[21] See, eg, Meerabux v Attorney-General of Belize [2005] EWCA Civ 634; [2005] 2 AC 513 at [25] (PC); Gillies v Secretary of State for Work and Pensions [2006] UKHL 2; [2006] 1 All ER 731 at [39] (Baroness Hale).

[22] [2000] HCA 63; (2000) 205 CLR 337.

[23] [2000] HCA 63; (2000) 205 CLR 337 at 348.

[24] [2006] UKHL 2; [2006] 1 All ER 731.

[25] Ibid, at [38]. See also Lord Hope at [23]. Canadian courts have made similar remarks about independence and impartiality. See Lorne Sossin, “The Uneasy Relationship Between Independence and Appointments in Canadian Administrative Law” in Grant Huscroft and Michael Taggart (eds), Inside and Outside Canadian Administrative Law (Toronto: Univ of Toronto Press, 2006) 50, 51-56.

[26] See, eg, Lawal v Northern Spirit [2003] UKHL 35; [2004] 1 All ER 187 (successful challenge to tribunal in which counsel acted sometimes as a part time recorder to the tribunal and sometimes as counsel appearing before the tribunal. Arrangements for part time recorders changed in light of this case); Brooke v Parole Board [2008] EWCA Civ 29 (successful challenge to parole board, due to the limited tenure of members and subjection of the board to binding ministerial directions).

[27] See, eg, R (Al-Hasan) v Secretary of State for the Home Department [2005] UKHL 13; [2005] 1 All ER 927 (HL) (prison governor who authorised a prisoner to be searched was too closely connected to the issue to determine impartially a disciplinary charge resulting from the prisoner's alleged non-compliance with the search); Scotland) [2008] UKHL 62; [2008] 1 WLR 2416 (Jewish judge who determined an asylum appeal from a Palestinian Muslim was held not biased by reason of her membership of the International Association of Jewish Lawyers and receipt of the organisation's newsletter, which contained many anti-Palestinian articles because the same journal contained a great range of articles and opinions on the middle East); Re Duffy [2008] UKHL 4 (members of a commission that determined the path of nationalist parades through Northern Ireland cities were held to suffer an irredeemable bias because they had a strong history of involvement in one side of the nationalist debate in Northern Ireland).

[28] Lord Hope appeared to favour a similar view in Gillies v Secretary of State for Work and Pensions [2006] UKHL 2; [2006] 1 All ER 731 at [23] when he stated that “[I]mpartiality consists in the absence of a predisposition to favour the interest of either side in a dispute. Therein lies the integrity of the adjudication system”. This passage suggests that the personal impartiality of a decision-maker is inextricably linked to a wider institutional framework. See also the decision of the European Court of Human Rights in Findlay v UK [1997] ECHR 8; (1997) 24 EHRR 221 at [73], though the Court spoke of the concepts of independence and “objective impartiality” and FB v Director of Immigration [2008] (HCAL51/2007, 5 December 2008, Saunders J) at [199] where Saunders J commented that a decision-maker's lack of independence could infect the independence of his or her judgment.

[29] William Lucy, “The Possibility of Impartiality” (2003) 25 Oxford Journal of Legal Studies 3.

[30] See, eg, Ebner v Official Trustee [2000] HCA 63; (2000) 205 CLR 337 at 373 where Kirby J stated that independence and impartiality were both constitutional requirements for the courts. Gaudron J adopted a similar view: 362-365. Kirby J adhered to this view in Smits v Roach (2006) 227 CLR 423 at 464-465.

[31] One example is the exception of waiver. If the bias rule had a constitutional basis in judicial proceedings it would almost certainly not be capable of waiver in courts but would be in bodies that were not constitutionally entrenched, such as tribunals and other administrative bodies. A radically different approach to waiver of bias, depending whether the issue arose in a court or tribunal, would clearly be undesirable. See Mark Aronson, Bruce Dyer and Matthew Groves, Judicial of Administrative Action (Sydney: Thomson Reuters, 4th edn, 2009), pp 702-703. A different view is taken in Bridgette Toy-Cronin, “Waiver of the Rule Against Bias” (2002) 9 Auckland University Law Review 850, 864 where it is argued that bias constitutes a jurisdictional error and cannot, therefore, be waived by the parties.

[32] See, eg, Dr Ip Kay Lo v Medical Council of Hong Kong [2003] 3 HKLRD 851 at 856 (Cheung JA).

[33] This proposition would also align the position of tribunals in the legal structure of the HKSAR to that of many other jurisdictions such as Canada and Australia. In those jurisdictions courts and tribunals are subject to differing principles but many of the principles applicable to courts may be extended to tribunals, or vice versa depending on the character of the tribunal in question. The important point, which may become prominent in the HKSAR, is that the exact position of tribunals within legal systems is often uncertain. In the HKSAR this uncertainty may be due in part to what one commentator has suggested is the “inherent indeterminacy” of the Basic Law: “Basic Law, Basic Politics: The Constitutional Game of Hong Kong” (2007) 37 HKLJ 503 at 543.

[34] A different view is taken in Benny Y.T. Tai, “Basic Law, Basic Politics: The Constitutional Game of Hong Kong” (2007) 37 HKLJ 503 at 503 where it is argued that because the Basic Law “shares the characteristics and functions of most other constitutions” that its “real meaning and significance” can be understood by applying “similar methods and approaches to those used when interpreting a constitution.” See also Michael Dowdle, “Constitutionalism in the Shadow of the Common Law” in Hualing Fu, Lison Harris and Simon Young (eds), Interpreting Hong Kong's Basic Law --The Struggle for Coherence (New York: Pal-grave McMillan, 2008, p 73. Dowdle complains of “common-law constitutional parochialism ... that needlessly constrains the migration of constitutional insight.” The implicit charge is that the transmission of constitutional ideas between jurisdictions should not be inhibited by the particular structural variations between jurisdictions. Such arguments do not preclude the possibility that significant differences may arise between different jurisdictions.

[35] This might occur in a federation, where the constitutions varied between provinces or states. In the Australian context, Kirby J has suggested that the constitutional basis of the bias rule extends to both federal and non-federal courts in Australia, even though the State courts do not have entrenched constitution and are not subject to the stringent version of the separation of powers that applies at the federal level: Smits v Roach (2006) 227 CLR 423 at 464-465 (citing Ebner v Official Trustee [2000] HCA 63; (2000) 205 CLR 337 at 362-363 where Gaudron J reasoned that the integrated nature of Australia's judicial system required a uniform requirement of judicial impartiality.) This view has not been accepted by a majority of the High Court of Australia.

[36] The importance of comparative jurisprudence to public law in the HKSAR is explained in Sir Anthony Mason, “The Place of Comparative Law in Developing the Jurisprudence of the Rule of Law and Human Rights in Hong Kong” (2007) 37 HKLJ 299.

[37] This reference to “constitutional structure” is not intended to indicate any disagreement with the statement of Li CJ that “The Basic Law is a national law and the constitution of the Region”: NgKaLing v Director of Immigration [1999] HKCFA 73; [1999] 1 HKLRD 315 at 337. It is instead intended to indicate that the Basic Law only can and should be understood within the wider political and legal environment of the HKSAR.

[38] Na Ka Ling v Director of Immigration [1999] HKCFA 73; [1999] 1 HKLRD 315 at 337.
[39] This statement is not intended to suggest that constitutional principles in general, or in Hong Kong in particular, are of a purely legal character. Non-legal or political principles are equally important but they may also be as nebulous and evolve over time in much the same way as the principles that arise from written constitutional documents and other fundamental instruments of government. See, eg, Sonny Shiu Hing-Lo, “The Mergence of Constitutional Conventions in the Hong Kong Special Administrative Region” (2005) 35 HKLJ 103, where it is argued that constitutional conventions may be assuming a new and important role in post-colonial Hong Kong. That author suggests that a form of ministerial responsibility is evolving in the HKSAR and that this doctrine is being fashioned largely by reference to the political relationship between the HKSAR and the Beijing government.

[40] Mason, n 35 above, p 305.

[41] See, eg, Peter Wesley-Smith, “Judges and Judicial Power Under the Hong Kong Basic Law” (2004) 34 HKLJ 83 and Peter Wesley-Smith, “Judicial Independence and the Rule of Law in Hong Kong” in S. Tsung (ed), Judicial Independence and the Rule of Law in Hong Kong (New York: Palgrave McMillan, 2001).

[42] (2003) 6 HKCFAR 476 at 489.

[43] Re Medicaments and Related Classes of Goods (No 2) [2000] EWCA Civ 350; [2001] 1 WLR 700 at [37]- [39] (CA).

[44] Apprehended bias has been variously referred to as “apparent”, “imputed”, “suspected” or “presumptive” bias: Anderton v Auckland City Council [1978] 1 NZLR 657 at 680 (SC NZ); Australian National Industries Ltd v Spedley Securities Ltd (in Liq) (1992) 26 NSWLR 411 at 414 (NSW CA); Re Medicaments and Related Classes of Goods (No 2) [2000] EWCA Civ 350; [2001] 1 WLR 700 at [38] (CA).

[45] It will be argued below, however, that the objective nature of the test for apprehended bias is often distorted by the courts.

[46] The extent to which evidence from the decision-maker (which usually takes the form of a denial of any bias) may be considered by an appellate court remains unsettled. The Privy Council has suggested that evidence of the deliberations of multi-member tribunals would never be admissible in an appellate consideration of bias: Roylance v General Medical Council (No 2) [1999] UKPC 16; [2000] 1 AC 311 at 323-5. Courts usually place no weight on statements by decision-makers that they are not biased. See, eg, Porter v Magill [2002] 2 AC 357 at 495 (Lord Hope); Locobail (UK) Ltd v Bayfield Properties Ltd [1999] EWCA Civ 3004; [2000] QB 451 at 477-478. See also Helow v Secretary of State for the Home Department (Scotland) [2008] UKHL 62; [2008] 1 WLR 2416 at [39] where Lord Cullen explained that evidence about a claim of bias “may include a statement from the judge as to what he or she knew at the time, although the court is not necessarily bound to accept any such statement at face value, there can be no question of cross-examining the judge on it, and no attention will be paid to any statement by the judge as to the impact of any knowledge on his or her mind.” It is also important to note that public policy considerations normally preclude the introduction of evidence about the deliberations of courts and tribunals: Cheung v Insider Dealing Tribunal [2000] 1 HKLRD 807. This rule also restricts the evidence that may be led in most claims of bias.

[47] See, eg, Sun v Minister for Immigration and Ethnic Affairs [1997] FCA 1488; (1997) 151 ALR 505 at 551-552 (Fed Ct, Aust); Gamaethige v Minister for Immigration and Multicultural Affairs [2001] FCA 565; (2001) 109 FCR 424 at 443 (Fed Ct, Aust). See also Porter v Magill [2002] 2 AC 357 at 489 where Lord Hope accepted that proof of actual bias was “likely to be very difficult”.

[48] This expression of the bias test was suggested by the English Court of Appeal in Re Medicaments and Related Classes of Goods (No 2) [2000] EWCA Civ 350; [2001] 1 WLR 700 at 711 and adopted by the House of Lords in Porter v Magill [2002] 2 AC 357. The Australian test, which is explained below, also adopts an objective assessment and will be satisfied if there is a “possibility” that the decision-maker might not be impartial: Ebner v Official Trustee [2000] HCA 63; (2000) 205 CLR 337 at 345.

[49] The fact that the court may uphold a claim of bias without making a direct finding against the court or judge in question enables a claim of bias to be determined without an adverse direct adverse finding a court or judge, which might undermine the very integrity that the rule against bias is supposed to maintain. See Webb v R [1994] HCA 30; (1994) 181 CLR 41 at 71-72 (Deane J).

[50] See, eg, R v Luskink Ex p Shaw (1980) 32 ALR 47 at 50 (Gibbs CJ, HCA).

[51] [1999] EWCA Civ 3004; [2000] QB 451.

[52] Ibid, at [3].

[53] Minister for Immigration and Multicultural Affairs Ex p Jia (2001) 205 CLR 507 at 541.

[54] [2000] HCA 48; (2000) 201 CLR 488.

[55] Ibid, at 517.

[56] [1993] UKHL 1; [1993] AC 646 at 670.

[57] [1994] HCA 30; (1994) 181 CLR 41.

[58] Ibid, at 71 (Deane).

[59] Ibid, at 52.

[60] Ibid.

[61] [1999] EWCA Civ 3004; [2000] QB 451.

[62] Ibid, at 477.

[63] [2000] EWCA Civ 350; [2001] 1 WLR 700.

[64] Ibid, at 726-727.

[65] [2002] 2 AC 357.

[66] (2003) 6 HKCFAR 322.

[67] (2003) 6 HKCFA 322 at 333. An important related point was that the CFA appeared unconvinced that the lower court had misapplied the bias test. The CFA ultimately concluded that, even if the bias test had been misapplied (a point it firmly declined to rule upon) that would not give rise to a question of great general or public importance sufficient to grant leave to appeal: 333.

[68] See, eg, Pacific China Holdings Ltd v Grant Pacific Holdings Ltd [2007] 3 HKLRD 742. See also PCCW-HKT Telephone Ltd v Telecommunications Authority [2008] 2 HKLRD 282 at 288 where the court noted that it was “common ground” between the parties that test espoused in Porter was the appropriate one. See also Jung v ZTE Corp [2008] 4 HKLRD 776 at 794 where it was noted that the Porter test has not been decisively endorsed by the CFA but the case had been applied in many other Hong Kong cases.

[69] See, eg, Rowse v Secretary for Civil Service [2008] HKCFI 549; [2008] 5 HKLRD 217 at 256.

[70] Deacons v White & Case Ltd Liability [2003] HKCA 601; [2003] 2 HKLRD 840 at [17].

[71] R v Lippé [1991] 2 SCR 114 at 152 (Lamer CJ, Sopinka and Cory JJ).

[72] Taylor v Williamsons (A firm) [2002] EWCA Civ 1380. English law enables courts in some circumstances to issue preliminary judgments to the parties. The reasoning of the Court of Appeal in the Taylor case suggests that although this course may be possible it may not necessarily be advisable. [FN72]. [2000] HCA 48; (2000) 201 CLR 488.

[73] [2000] HCA 48; (2000) 201 CLR 488.

[74] Ibid, at [53]. A similar formula to that developed Kirby J was adopted a few years earlier in RDS v R [1997] 3 SCR 484 at [36] where L'Heureux-Dube and McLachlin JJ suggested that bias should be “evaluated through the eyes of the reasonable, informed, practical and realistic person who considers the matter in some detail ... The person postulated is not a ‘very sensitive or scrupulous' person, but rather a right-minded person familiar with the circumstances of the case.”

[75] Judicial citations of Kirby's statement in Hong Kong include PCCW-HKT Telephone Ltd v Telecommunications Authority [2008] 2 HKLRD 282 at [16]; Jung Science Information Technology Co Ltd v ZTE Corp [2008] 4 HKLRD 776 at [52]. Kirby J's statement is also frequently cited in England. See, eg, Lawal v Northern Spirit [2003] UKHL 35; [2004] 1 All ER 187 at [14] (HL); R v Abdroikov [2007] 1 WLR 2697 at [15] (HL).

[76] There are exceptions. Cases in the HKSAR that have quoted the full passage of Kirby J noted in the text of this article include: Pacific China Holdings Ltd v Grand Pacific Holdings Ltd [2007] HKLRD 741 at 749 (Burrell J); Chong Wai Lee Charles v Insider Dealing Tribunal [2006] HKCFI 38 at [53] (Reyes J).

[77] Ibid, at [53].

[78] Johnson v Johnson [2000] HCA 48; (2000) 201 CLR 488 at 506. See also R v Abdroikov [2007] 1 All ER 315 at [81] where Lord Mance conceded that the “fair-minded and informed observer” was “in large measure the construct of the court.”

[79] (2006) 227 CLR 423 at 457 [96].

[80] Basten JA of the New South Wales Court of Appeal seemed mindful of similar concerns when he explained that there was no harm in attributing detailed knowledge to the reasonable fair-minded observer “so long as it is not an excuse for fuzzy thinking about the test to be applied:” Lee v Bob Chae-Sang Cha [2008] NSWCA 13 at [43] (Basten JA, Hodgson and Bell JJA agreeing).

[81] (2006) 227 CLR at 457 [97].

[82] Raybos Australia Pty Ltd v Tectran Corp Pty Ltd (1986) 6 NSWLR 272 at 276. See also Commissioners of Corrective Services v Government and Related Appeal Tribunal [2004] NSWCA 291 at [25] (Giles JA, Sheller and Ipp JJA agreeing).

[83] Taylor v Lawrence [2002] EWCA Civ 90; [2003] QB 528.

[84] Ibid, at 548-549.

[85] Jung Science Information Technology Co Ltd v ZTE Corp [2008] 4 HKLRD 776 at [55].

[86] [2006] UKHL 2; [2006] 1 All ER 731 at [39].

[87] Simon Atrill, “Who is the ‘Fair-Minded and Informed Observer’? Bias after Magill ” (2003) 62 Cambridge Law Journal 279 at 283.

[88] [1852] EngR 789; (1852) 3 HLC 759.

[89] See above, nn 15-16.

[90] Many of these cases are reviewed by Kirby J in Ebner v Official Trustee [2000] HCA 63; (2000) 205 CLR 337 at 376-387.

[91] [1999] UKHL 1; [2000] 1 AC 119 . The initial extradition appeal is reported at [2000] 1 AC 61 . The wider circumstances of the Pinochet cases are examined in detail in Diana Woodhouse (ed), The Pinochet Case: A Legal and Constitutional Analysis (Oxford: Hart Publishing, 2000).

[92] [1999] UKHL 1; [2000] 1 AC 119 at 132-133.

[93] It also removed the need to examine other difficult arguments, such as the relevance of Lord Hoffman's fund raising activities for Amnesty or the fact that Lady Hoffman was also very involved in the work of Amnesty.

[94] A point conceded by Amnesty: [1999] UKHL 1; [2000] 1 AC 119 at 135.

[95] Locobail (UK) Ltd v Bayfield Properties Ltd [1999] EWCA Civ 3004; [2000] QB 451. The description of the case as “dogmatic” is taken from Michael Taggart, “Administrative Law” [2003] New Zealand Law Review 99 at 101.

[96] Membership of Masonic associations by senior public officials has been a very contentious issue in England in recent years.

[97] [2000] HCA 63; (2000) 205 CLR 337.

[98] The shares were clearly worth far less in relative terms in comparison to the enormously valuable holding of the Lord Chancellor in Dimes. In the Ebner case the judge was a contingent beneficiary and director of a family trust that held around 9,000 bank shares. The bank was not a direct party to the case (a bankruptcy proceeding) but had a pecuniary interest in its outcome. The judge disclosed the interest and rejected a recusal request from the bankrupt party. In the co-joined Clenae case the judge inherited 2,400 shares in a bank that was party to proceedings before him. The inheritance occurred while the decision was reserved. The judge did not disclose the inheritance to the parties and later gave judgment for the bank. In Ebner the shares were worth about AUD $30,000. In Clenae they were worth about AUD $100,000. Both holdings were in the same bank, which had a market value at that time of about AUD$8billion. In each case there was no clearly possibility that any judgment of the court could influence the value of the bank concerned or either parcel of shares in issue.

[99] Ebner v Official Trustee [2000] HCA 63; (2000) 205 CLR 337 at 356 (Gleeson, McHugh, Gummow and Hayne JJ, Callinan J agreeing).

[100] Ibid, at 355-357 (Gleeson, McHugh, Gummow and Hayne JJ, Callinan J agreeing). Kirby J dissented, rejecting the reasoning of the majority as “an ahistorical interpretation” of Dimes that was not supported by either Dimes or later cases in which it was applied: at 378.

[101] Ebner v Official Trustee [2000] HCA 63; (2000) 205 CLR 337 at 356-357 (citations omitted).

[102] Ibid, at 345.

[103] See also Hot Holdings Pty Ltd v Creasy [2002] HCA 51; (2002) 210 CLR 438 at 447 (Gleeson CJ).

[104] In Hot Holdings Pty Ltd v Creasy [2002] HCA 51; (2002) 210 CLR 438 a clear majority of the High Court held that an applicant had not managed to articulate this connection, while the lower court had unanimously held to the contrary.

[105] [2007] NZCA 334; [2007] 3 NZLR 495.

[106] Ibid, at [42]. This reasoning was supported by Philip Joseph, Constitutional and Administrative Law in New Zealand (Wellington: Thomson Brookers; 3rd edn, 2007), p 994.

[107] Ibid.
[108] Flaherty v National Greyhound Racing Club Ltd [2005] EWCA Civ 1117 at [28], citing R v Inner West London Coroner Ex p Dallaglio [1994] 4 All ER 139 at 151. See also Imperial Oil Ltd v Attorney-General of Quebec (2003) 231 DLR (4th) 577 at [28] where the Supreme Court of Canada stated that a decision maker “must approach the issue submitted to him or her with an open mind, not influenced by personal interests or outside pressure” and Gillies v Secretary of State for Work and Pensions [2006] UKHL 2; [2006] 1 All ER 731 at [23] where Lord Hope explained that “[I]mpartiality consists in the absence of a predisposition to favour the interests of either side in the dispute.”

[109] Liteky v United States 510 US 540 at 550 (1994) (emphasis in original).

[110] William Lucy, “The Possibility of Impartiality” (2003) 25 Oxford Journal of Legal Studies 3 at 14.

[111] Ibid, 15.

[112] (1997) 151 DLR (4th) 193.

[113] Ibid, at [39].

[114] Ibid, at [29].

[115] Ibid, at [35] citing the Canadian Judicial Council, Commentaries on Judicial Conduct (Cowansville, Quebec, 1991) p12. L'Heureux-Dube J subsequently took this reasoning a step further in an extrajudicial speech in which she suggested that judges ought not to aspire to neutrality and should instead seek to identify and correct inequalities in life: “Reflections on Judicial Independence and the Foundations of Equality” (1999) 7 Canadian Journal of International Law Yearbook 95. This speech and her Honour's judgement in the RDS case provoked considerable controversy in Canada. The wider issues raised by the approach of L'Heureux-Dube J are considered thoughtfully in Regina Graycar, “Gender, Race, Bias and Perspective: OR, How Otherness Colours Your Judgment” (2008) 15 International Journal of the Legal Profession 73.

[116] [2007] 1 WLR 2679.

[117] The cases arose after legislative amendments reversed a longstanding prohibition upon police officers serving as jurors. Many common law jurisdictions maintain this prohibition.

[118] [2007] 1 WLR 2679 at [81].

[119] See, eg, R v Pintori [2007] EWCA Crim 1700; R v Ingleton [2007] EWCA Crim 2999; Vella v State of Western Australia [2007] WASCA 59; R v Goodall [2007] VSCA 63; (2007) 15 VR 673 (Vic CA).

[120] [2002] 2 AC 357.

[121] See Saima Hanif, “The Use of the Bystander Test for Apparent Bias” (2005) 10 Judicial Review 78 and Simon Atrill, “Who is the ‘Fair-Mind and Informed Observer’? Bias After Magill ” (2003) 62 Cambridge Law Journal 279. 39 Hong Kong L. J. 485